Treacher Collins Syndrome

Bilateral, symmetrical mandibulofacial dysostosis of cranial neural crest origin, caused by variants in TCOF1, POLR1D, POLR1C or POLR1B, genes involved in ribosome biogenesis; includes the allelic disorder of POLR1C (hypomyelinating leukodystrophy) and the differential diagnosis of the other facial dysostoses.

Literature and regulatory information updated: 10 October 2026.

0. Initial Reference Table

1. Introduction

1.1. Founding clinical descriptions

Treacher Collins syndrome (TCS) is the most frequent mandibulofacial dysostosis and serves as the reference for the whole group of facial dysostoses [23]. The first case is attributed to Thomson, in 1846, and in 1889 Berry described two cases with a congenital defect of the lower eyelids, including a mother and daughter, which suggested a hereditary basis for the first time [2, 24]; nevertheless, the disorder became associated with the English ophthalmologist Edward Treacher Collins, who in 1900 described two patients with "congenital symmetrical notches in the outer part of each lower eyelid and defective development of the malar bones" [2, 4]. In the 1940s, Adolphe Franceschetti and David Klein gathered extensive descriptions, proposed the name mandibulofacial dysostosis and in 1949 published the monograph that established the clinical picture, with a classification into five forms (complete, incomplete, abortive, unilateral and atypical); hence the eponyms Franceschetti-Klein syndrome and Franceschetti-Zwahlen-Klein syndrome, the latter more widely used in the European literature [1, 2, 24]. Treacher Collins's original text and the monograph by Franceschetti and Klein are not indexed in the bibliographic databases consulted, and they are cited through the reviews that summarise them [2, 4].

1.2. Milestones of molecular discovery

In 1996, the Treacher Collins Syndrome Collaborative Group mapped the gene for the autosomal dominant form to 5q32-q33.1, isolated by positional cloning a gene of then unknown function, which it named Treacle (now TCOF1), and demonstrated different mutations in five unrelated families, all creating a premature termination codon [25]. The Brazilian cohort of Splendore et al. (2000), with 28 families, raised the TCOF1 detection rate to 93% and showed the clustering of mutations in certain exons, with no genotype–phenotype correlation [26]. In 2011, Dauwerse et al. identified, in a 3-year-old boy negative for TCOF1, a 156 kb de novo deletion at 13q12.2 encompassing the whole of POLR1D and exon 1 of LNX2 [7]. POLR1D is the gene for a subunit shared by RNA polymerases I and III. The same authors subsequently found 20 heterozygous POLR1D mutations in 252 individuals and biallelic POLR1C mutations in three, which established genetic heterogeneity and strengthened the ribosomopathy hypothesis [27]. In 2014, Schaefer et al. described the first autosomal recessive form due to POLR1D, with the homozygous variant c.163C>G p.(Leu55Val) in two consanguineous Turkish families, in which the parents and an unaffected sister were heterozygous [7, 28]. In 2020, Sanchez et al. used exome sequencing to identify POLR1B variants in patients negative for the known genes and defined TCS type 4 [13]. More recently, long-read sequencing resolved a retrotransposon insertion in intron 17 of TCOF1 in a case that had remained undiagnosed for almost a decade [29].

1.3. Nomenclature and classification

GeneReviews lists mandibulofacial dysostosis and Treacher Collins-Franceschetti syndrome as synonyms; the autosomal dominant form has also been called Franceschetti-Zwahlen-Klein syndrome and zygoauromandibular dysplasia [1]. The Nosology of Genetic Skeletal Disorders (2023 revision) places it in the craniofacial dysostoses group with dyadic gene-based names, "TCOF1-, POLR1B-, POLR1C- or POLR1D-related mandibulofacial dysostosis" [1, 3]. OMIM numbers four phenotypic entries by gene: TCS1 (154500, TCOF1), TCS2 (613717, POLR1D), TCS3 (248390, POLR1C) and TCS4 (618939, POLR1B); Orphanet groups them into a single entity, ORPHA:861, classified as a malformation syndrome and included in the mandibulofacial dysostoses group (ORPHA:155899) [1, 5, 14]. The ERN CRANIO European consensus addresses TCS together with Nager and Miller syndromes under the name facial dysostosis syndromes (FDS), because they share the multidisciplinary approach [20].

1.4. Definition of the entities in this article

TCS is defined by a bilateral, symmetrical oto-mandibular dysplasia, with malar hypoplasia, micrognathia, downslanting palpebral fissures, lower eyelid coloboma, microtia and hearing loss, without limb anomalies and with normal intelligence [5]. TCS1 corresponds to the presence of a heterozygous pathogenic or likely pathogenic variant in TCOF1, or of a 5q32-q33.1 deletion including TCOF1; TCS2 to a heterozygous, or more rarely biallelic, variant in POLR1D; TCS3 to biallelic variants in POLR1C; and TCS4 to a heterozygous variant in POLR1B [1]. The 5q32 deletions that remove TCOF1 together with CAMK2A and other neighbouring genes produce TCS with intellectual disability and are included in TCS1 as a contiguous gene syndrome [30, 31]. Hypomyelinating leukodystrophy due to biallelic POLR1C variants is an allelic disorder without a TCS-type craniofacial phenotype and is covered in section 2 and in the differential diagnosis [15, 17]. The other mandibulofacial dysostoses (Nager syndrome, Miller syndrome, mandibulofacial dysostosis with microcephaly, Cincinnati-type acrofacial dysostosis) are included only as differential diagnoses; haematological ribosomopathies, such as Diamond-Blackfan anaemia, are mentioned only as biological context [1, 32].

2. Molecular Genetics and Pathophysiology

2.1. Ribosome biogenesis and the cranial neural crest

The skeleton and connective tissue of the face derive from the mesenchyme produced by cranial neural crest cells, which migrate into the first and second pharyngeal arches [1, 32]. The four TCS genes are expressed in the neural crest, their products colocalise in the nucleolus and they take part in transcription of ribosomal DNA (rDNA) [1, 11]. RNA polymerase I (Pol I) synthesises the precursor of the large ribosomal RNAs (rRNA), Pol II the messenger RNAs and Pol III the transfer RNAs (tRNA) and 5S rRNA [32]. The 47S/45S pre-rRNA transcribed by Pol I is processed into 18S, 5.8S and 28S rRNA; treacle interacts with upstream binding factor (UBF), recruits Pol I, UBF and Nopp140 to the rDNA promoter and takes part in 2′-O-methylation of pre-rRNA [13, 33, 34]. Reduced function of any of these components decreases the production of mature ribosomes in the neuroepithelium and neural crest [1, 27].

The Tcof1+/− mouse model showed that haploinsufficiency causes a deficiency of migrating neural crest cells and severe craniofacial malformations, and that treacle is required cell-autonomously for neural crest formation and proliferation [35]. Apoptosis is increased in the neural folds before their fusion [36]. Defective ribosome biogenesis stabilises the p53 protein, which activates cyclin G1-mediated cell-cycle arrest and neuroepithelial apoptosis; genetic or chemical inhibition of p53 prevents apoptosis and corrects the craniofacial anomalies in the mouse, without restoring ribosome biogenesis [37]. Oxidative stress contributes to DNA damage and to the death of neuroepithelial cells, and maternal antioxidant supplementation attenuates or prevents the craniofacial anomalies in the Tcof1+/− mouse [38]. These data explain the structural, non-progressive nature of the phenotype: it is a ribosomopathy expressed in a specific cell lineage and developmental window [31]. Treacle also has a second function, dependent on monoubiquitination by the BCR(KBTBD8) complex: in association with NOLC1, it remodels the translational programme in favour of neural crest fate [31].

Calo et al. (2018) showed that TCS-associated genetic perturbations displace the helicase DDX21 from the nucleolus to the nucleoplasm, inhibit rRNA processing and activate p53; cranial neural crest cells become sensitised to apoptosis, and preventing the nucleolar loss of DDX21 corrects the susceptibility to apoptosis and the craniofacial phenotype in the models studied; rDNA damage has tissue-selective, dose-dependent effects, shared with the ribosomal perturbations of Diamond-Blackfan anaemia [6]. A 2026 preprint, not peer reviewed, proposes that neural crest cells transitioning to mesenchymal specification are extremely sensitive to TCOF1 dosage, but not to the dosage of Pol I itself, and that TCOF1 depends on neural crest-specific enhancers that buffer fluctuations in promoter factors; this model would explain the cellular origin of the disorder and the predominance of TCOF1 among the causes [39].

2.2. TCOF1 and treacle

TCOF1 is located at 5q32-q33.1 and has 27 exons, including exons 6A (231 bp) and 16A (108 bp), identified after the initial cloning [33, 40]. The classical treacle isoform is a nucleolar phosphoprotein of 1411 amino acids and about 144 kDa; the transcript containing exon 6A encodes a protein 77 amino acids longer, and Antal et al. report 152 kDa, a difference that reflects the isoform considered [33, 40]. The current MANE Select transcript, NM_001371623.1, encodes the 1489-amino-acid isoform (NP_001358552.1) [10, 41]. The protein has an N-terminal domain with a nuclear export signal and a LisH motif, a central domain with 11 repeats rich in phosphorylation sites for casein kinase 2 and protein kinase C, and a C-terminal domain with several nuclear localisation signals (NLS) in exons 23, 24 and 25 [33, 40]. Treacle shuttles between the nucleolus and the cytoplasm; its reduction displaces UBF and inhibits rRNA synthesis; it is also a component of the NBS1 complex and takes part in the DNA damage response [33]. The disease mechanism is loss of function through haploinsufficiency [1, 4].

2.3. POLR1D, POLR1C and POLR1B

POLR1D (13q12.2; 3 exons) encodes RPAC2 (AC19), a subunit shared by Pol I and Pol III; Pol III transcribes 5S rRNA, tRNAs, small nuclear RNAs and microRNAs [31, 33]. In yeast, TCS-associated mutations destabilise the integrity of the Pol I and Pol III complexes [42]. The heterozygous variant acts through haploinsufficiency (AD form) and biallelic loss of function produces the AR form, in a dose-dependent mechanism; ClinGen assigns POLR1D a haploinsufficiency score of 3 (sufficient evidence) [11, 31]. The index patient with homozygous POLR1D studied by Schaefer et al. had a 50% reduction in TCOF1 mRNA, a sign of interdependence between the components of the pathway [28].

POLR1C (6p21.1; 9 exons) encodes RPAC1 (AC40), a core subunit shared by Pol I and Pol III, which forms a heterodimer with RPAC2 and is part of the structure that closes the DNA-binding cleft [31, 33]. In TCS3, the mechanism is biallelic functional depletion [27, 33]. The heterodimerisation interface between POLR1D and POLR1C involves residues 39–112 of POLR1D [41].

POLR1B (2q14.1; 15 exons) encodes RPA2, the second largest subunit and a subunit exclusive to Pol I [1, 33]. The affected residues, Ser682 and Arg1003, form hydrogen bonds that stabilise the Ser682–Gln724 α-helix, linked to the "bridge helix" of POLR1A in the catalytic centre; the proposed mechanism is loss of function [13]. Knockdown of polr1b in zebrafish reproduces a TCS-like craniofacial phenotype, with deficient Pol I-dependent rRNA, massive p53-associated apoptosis in the neuroepithelium and reduction of neural crest derivatives [13]. In yeast, mutations homologous to those of TCS4 reduce growth, accumulate Pol I in the 5′ region of the rDNA and increase sensitivity to the Pol I inhibitor BMH-21 [43].

Table sources: [1, 12, 13, 26, 31, 33].

2.4. Spectrum of TCOF1 variants

More than 200 pathogenic TCOF1 variants have been published, mostly small deletions of 1 to 40 nucleotides that create a premature termination codon and lead to nonsense-mediated mRNA decay (NMD) or to a truncated protein [40]. In an aggregated systematic review of 50 publications from 19 countries (2000–2025), with 303 variant entries in 275 individuals, frameshift variants accounted for 66.01%, nonsense variants for 10.56%, missense variants for 10.23%, large deletions for 2.64% and splicing variants for 2.31% [4]. The review by Ulhaq et al., with 338 TCOF1 cases in 43 publications, found deletions in 66%, substitutions in 23%, insertions in 6%, duplications in 3% and indels in 2%; large deletions accounted for 4.4% [40, 44]. Hotspots in exons 10, 15, 16, 23 and 24, which concentrate more than half of the mutations, have been proposed, but the only true hotspot is exon 24, with 17.75% of the variants [40, 44].

The best-known recurrent variant is a 5 bp deletion in exon 24. In the historical nomenclature, still used by GeneReviews, it is written NM_001135243.2:c.4369_4373del p.(Lys1457GlufsTer12), and in older descriptions "c.4135delGAAAA"; in the MANE transcript it corresponds to NM_001371623.1:c.4372_4376del, NP_001358552.1:p.(Lys1458GlufsTer12), with the genomic coordinate NC_000005.10:g.150398380_150398384del, confirmed by VariantValidator as identical in both descriptions [1, 10, 40]. The difference of three nucleotides and one amino acid results solely from the choice of transcript; transcript NM_000356.4, of the 1411-amino-acid isoform, does not contain position c.4369 [10]. The variant accounts for 25% of exon 24 variants in the review by Ulhaq et al.; the Oxford Desk Reference states that about 17% of patients have a 5-nucleotide deletion at the 3′ end of the gene, and Teber et al. did not find it among 28 mutations in their German series [40, 45, 46]. The deletion was described for the first time in an African family in the 2026 South African cohort (ClinVar record SCV000597428.1) [11].

The protein descriptions published in four recent articles diverge from the VariantValidator normalisation, which prevails in the table; the variants of the Hungarian series were described on the non-MANE transcripts NM_001135243 (TCOF1) and NM_001374407 (POLR1D) and are reproduced with the article's transcript: TCOF1 NM_001135243.2:c.1371_1372insT p.(Lys458*), NM_001135243.2:c.2103_2106del p.(Ser701Argfs*9) and NM_001135243.2:c.4369_4373del, and POLR1D NM_001374407.1:c.295G>C p.(Gly99Arg); in the MANE transcript they correspond to NM_001371623.1:c.1371_1372insT p.(Lys458Ter), NM_001371623.1:c.2103_2106del p.(Ser701ArgfsTer9), NM_001371623.1:c.4372_4376del p.(Lys1458GlufsTer12) and NM_015972.4:c.295G>C p.(Gly99Arg), the latter classified as a variant of uncertain significance (VUS) [10, 40].

2.5. Variants of POLR1D, POLR1C and POLR1B

The hotspots described lie in exon 3 of POLR1D, exon 8 of POLR1C and exon 15 of POLR1B [44]. By 2019, five families with AR inheritance due to POLR1D or POLR1C had been published; Ghesh et al. added two sisters with mild TCS and compound heterozygous POLR1C variants [49]. Dauwerse et al. identified five missense POLR1C variants in three patients with TCS [1, 27]. In POLR1B, all pathogenic variants reported up to 2021 clustered at two residues, Ser682 and Arg1003; p.(Arg1003Cys) is recurrent, and the four patients with this variant described by Enomoto et al. had external auditory canal atresia and microtia [13, 48].

2.6. Deletions, structural variants and non-coding variants

TCOF1 deletions range from a single exon (exon 3, 3.367 kb) to the whole gene [1, 33]. Bowman et al. studied 182 patients with a clinical diagnosis of TCS: 92 (50.5%) had a TCOF1 sequence variant (57% frameshift or initiation codon, 23% nonsense, 16% splicing and 4% missense) and 5.2% had an intragenic deletion, which suggests that their frequency is underestimated in series based on sequencing alone [1, 6, 50]. 5q32 deletions encompassing TCOF1 and CAMK2A, sometimes with SLC6A7, cause TCS with intellectual disability; in the French series, 2 of the atypical patients with intellectual disability or microcephaly had this type of deletion [12, 30, 31]. The largest insertion, c.484_668ins185bp in exon 5 (published designation, inserted sequence not specified), has been described in twin girls [33]. In 2026, the combination of fibroblast RNA sequencing (expression outlier analysis), optical genome mapping and long-read DNA and RNA sequencing identified a 3396 bp insertion in intron 17 of TCOF1, an SVA (SINE-VNTR-Alu) retrotransposon hybrid of subtypes C and E; the element is partially exonised (about 750 bp), generates frameshifted transcripts subject to NMD and shifts expression towards a short isoform lacking a nucleolar localisation signal, with a long-to-short isoform ratio of 5:19 on the mutant allele and 29:1 on the normal allele [29].

2.7. Genotype–phenotype correlations

TCS due to TCOF1 tends to be more severe than that caused by the POLR1 genes, although the data are limited; external auditory canal atresia, downslanting fissures and lower eyelid coloboma are more frequent in TCOF1-related TCS [1, 44]. Within TCOF1, duplications are associated with milder presentations, the 5 bp exon 24 deletion with greater severity than other variants in the same exon, and exon 15 variants with a significantly lower frequency of microtia, conductive deafness and canal atresia [1, 44]. Ulhaq et al. described greater severity in Asian patients with TCOF1 variants than with POLR1 variants [44]. In the 2026 aggregated review, truncating variants clustered in exons 11 to 24, frameshifts in exons 23–24 and deletions of exons 2–7 were more frequent among severe cases, but early frameshifts and 5′ deletions also produced severe phenotypes, so the position of the variant does not predict severity in an individual [4]. The French series of 146 patients found no genotype–phenotype correlation, nor did the 2000 Brazilian cohort [12, 26].

2.8. Penetrance, expressivity, mosaicism and modifiers

Penetrance is high, but reduced penetrance has been documented for TCOF1 and POLR1D, particularly with the recurrent exon 24 deletion [1, 40, 51]. In a Hungarian family, the c.4369_4373del variant (historical nomenclature) was transmitted by several asymptomatic female carriers to two affected second cousins, and the POLR1D variant c.295G>C was inherited from a mother without signs [40]. Expressivity is highly variable, even within the same family: some carriers have features so subtle that they go undiagnosed, while others have life-threatening airway obstruction [1, 2, 52]. The apparent anticipation observed in Brazilian families was due to ascertainment bias [26]. Two Hutterite sisters with unaffected parents, described in 1985 as a possible recessive form, in fact had a heterozygous 4 bp TCOF1 deletion also present in their unaffected father, an example of incomplete penetrance [6].

Parental mosaicism was confirmed or strongly suspected in 6 of 275 individuals (2.18%) in the 2026 aggregated review, mostly paternal; in all of them, the offspring had moderate to severe TCS [4]. In family 2 of that study, the clinically unaffected father had the c.3054_3060dup variant in about 10.84% of reads in blood and 8.74% in semen, detected by high-depth NGS [4]. In TCS4, a father with mild dysmorphism had about 16% mosaicism for POLR1B p.(Arg1003Ser) [13]. In 10 sporadic cases with determined parental origin, the mutation was paternal in 7 and maternal in 3, with no detectable effect of parental age; Li et al. cite that up to 80% of de novo mutations arise on the paternal haplotype [4, 6, 53].

Variability is modulated by factors beyond the variant. In Tcof1+/− mice on different genetic backgrounds, the combination of endogenous treacle levels and reactive oxygen species (ROS) correlates with severity; the genotype determines craniofacial shape and the redox state determines bone size [54].

2.9. Allelic disorder: POLR1C-related hypomyelinating leukodystrophy

Different biallelic POLR1C variants cause a distinct disorder, hypomyelinating leukodystrophy 11, within the spectrum of POLR3-related leukodystrophy [15, 31]. The picture includes cerebellar ataxia, dystonia, spasticity, abnormal dentition, hypogonadotropic hypogonadism, short stature in about half of patients, sometimes with growth hormone deficiency, and severe progressive myopia, with a hypomyelinating pattern on magnetic resonance imaging (MRI) and relative sparing of specific structures [31, 32]. POLR1C accounts for about 5% of cases of POLR3-related leukodystrophy and is tested when POLR3A and POLR3B are negative [15]. Thiffault et al. described eight cases and showed that leukodystrophy variants, such as NM_203290.4:c.95A>T p.(Asn32Ile) and NM_203290.4:c.221A>G p.(Asn74Ser), impair the assembly and nuclear import of Pol III, but not of Pol I, with reduced binding to Pol III target genes; TCS variants do not have this effect [15, 17]. Patients with TCS3 have no evidence of demyelination [1]. A 2026 preprint described, in the polr1c mutant zebrafish, reduced myelination, Tp53 activation in oligodendrocyte precursors without apoptosis, and reduced transcription of rRNA by Pol I and of some tRNAs by Pol III; tp53 inhibition reduced cell death without restoring myelination [55]. On 2 February 2026, ClinGen classified the relationship between POLR1C and POLR1C-related disease as definitive, with AR inheritance [31]. No other allelic phenotypes are known for germline variants of TCOF1, POLR1B or POLR1D [1].

2.10. Actionability

ClinGen has not issued an actionability assertion (adult or paediatric) for TCOF1 or POLR1C; paediatric management, in particular neonatal airway planning and hearing habilitation, is actionable, but the genes have not gone through that process [31].

3. Clinical Spectrum and System-Specific Manifestations

3.1. General picture and variability

TCS is characterised by bilateral, symmetrical hypoplasia of the zygomatic bones, maxilla and mandible, with downslanting palpebral fissures, a convex facial profile and a prominent nose, lower eyelid abnormalities, absent, small, malformed or posteriorly rotated ears, and atresia or stenosis of the auditory canals [1]. Displacement of the hair towards the preauricular region and the cheek is a characteristic sign [1, 2]. Severity ranges from barely perceptible features to profound mandibular hypoplasia with fatal neonatal respiratory compromise, and variability is wide within the same family [2, 4]. More than 5000 individuals with a pathogenic variant in one of the four genes have been identified [1]. Intelligence is typically normal; intellectual disability is described in fewer than 5% of cases and should prompt consideration of a contiguous deletion or another diagnosis [1, 2].

The GeneReviews frequencies derive from the series of Splendore et al., Teber et al. and Vincent et al. The TCOF1 column shows the ranges compiled by Antal et al.

Table sources: [1, 40]. In the 70 patients with TCS1 in the French series, the frequencies were: downslanting fissures 100%, malar hypoplasia 99%, conductive deafness 91%, mandibular hypoplasia 87%, external auditory canal atresia 72%, microtia 71%, lower eyelid coloboma 65%, asymmetry 53%, projection of hair onto the cheek 48%, cleft palate 22%, choanal stenosis or atresia 14% and cardiac malformation 12%; brain, renal and limb anomalies were rare [6, 12]. The GeneReviews summary gives conductive hearing loss in 40%–50%, a figure also given by Smith's, whereas the table in the same GeneReviews places it above 75%; series with systematic audiological assessment give 83%–100%; the sources do not explain the discrepancy, which may depend on the assessment method and the definition of hearing loss [1, 2, 19, 40].

3.2. Cohort data and what they measure

The 2026 aggregated review brought together 275 individuals with a TCOF1 variant (32 fetuses or newborns; the oldest adult aged 50 years): 68 (24.73%) had severe disease (score ≥9 on the Teber and Vincent scales), 263 (95.64%) survived and 12 (4.36%) died shortly after birth from critical airway compromise or had a termination of pregnancy; conductive deafness was present in 200 (72.73%), cleft palate in 29 and intellectual or developmental disability in 39 [4]. Because it derives from case reports, the sample over-represents severe phenotypes and does not estimate the natural history of a population [4]. The multicentre cohort from four craniofacial units (Rotterdam, London, Boston and Toronto) studied 248 patients (240 confirmed), and the French series 146 patients with molecular study [12, 56].

3.3. Eyes and adnexa

In the retrospective multicentre cohort of 194 patients (Rotterdam, London, Boston and Toronto), primary ocular anomalies were present in 98.5%: downslanting fissures in 93.8%, lower eyelid colobomas in 69.6% and total or partial absence of the lower eyelashes in 42.8%; secondary anomalies affected 34.5% (epiphora 24.2%, exposure keratopathy 14.4%), strabismus 27.3%, refractive errors 49.5% and visual impairment 4.6% [1, 57]. Only 49.5% of patients had been seen by an ophthalmologist or optometrist, and the first visual acuity was measured at a mean age of 6.96 years; there was no association between facial severity and visual impairment, and ocular anomalies were present even in mild cases [57]. The coloboma appears as a V-shaped notch at the junction of the medial two-thirds and the lateral third of the lower eyelid [23]. In the Philadelphia series of 24 patients (1980–1991), there was visual loss in 37%, amblyopia in 33%, significant refractive errors in 58%, anisometropia in 17% and strabismus in 37%; visual loss is mainly due to amblyopia secondary to refractive errors, anisometropia, strabismus or ptosis, and amblyopia was more resistant to treatment than usual [2, 58]. Another series cited by Plomp et al. found bilateral absence of the lower lacrimal puncta in 36%, absent eyelashes in 50%, scleral show in 50%, refractive errors in 86% and absence of the lateral canthal tendon in 64% [19]. In the Children's Hospital of Philadelphia cohort (2009–2023; 50 children, mean follow-up 10.5 years), downslanting fissures were present in 100%, eyelash deficiency in 70%, coloboma in 54% and significant refractive error in 50%; coloboma was associated with exposure keratopathy (30% versus 4%), and, in the 78 eyes of 39 patients with recorded acuity, vision was good in 90% [59]. Upper eyelid coloboma, dacryostenosis, microphthalmia, cataract and ptosis occur occasionally [2].

3.4. Ear and hearing

The external and middle ear are the most affected and the inner ear is relatively spared, which is explained by the dual embryological origin of these structures; higher grades of microtia are associated with more severe malformations of the canal and middle ear [1, 60]. According to the review by Plomp et al., 93%–96% of patients have unilateral or bilateral conductive hearing loss and 7% mixed hearing loss; the degree is moderate (41–55 dB) to moderately severe (56–70 dB), the latter in up to 65% [19]. On computed tomography (CT), the external auditory canal is normal in 0%–15%, stenotic in 28%–31% and atretic in 54%–72%; the ossicles are hypoplastic or ankylosed in 33%–82% or absent in 22%–67%, mainly the malleus and incus; the inner ear is normal in 78%–100%, the mastoid is rarely pneumatised and the facial nerve may have an aberrant course [2, 19]. In a Chinese series of 14 patients studied with high-resolution CT, there was displacement of the facial nerve in 92.31%, a high jugular bulb in 61.54%, a low-lying middle fossa in 53.85%, oval window anomalies in 76.92% and canal stenosis with cholesteatoma in 30.77% [61]. Sensorineural hearing loss should also be screened for; the Oxford Desk Reference describes the deafness as mixed or sensorineural in one of its entries, a statement that contradicts the series with CT and audiometry, in which conductive hearing loss predominates [19, 23, 62]. About 95% of patients report hearing impairment and about half communication difficulties [19].

3.5. Airway and sleep

Airway obstruction may be present from birth, due to choanal atresia or stenosis or to severe micrognathia with glossoptosis; 12% to 18% of those affected have critical airway problems at birth requiring surgical intervention, and neonatal death is usually associated with airway obstruction [1, 2]. Mandibular hypoplasia measured by CT correlates with neonatal intubation or tracheostomy (r=0.519), with the need for a nasogastric tube or gastrostomy (r=0.355) and with cleft lip or palate (r=0.349) [63]. The obstruction is multilevel, from the nasal septum to the trachea, and most marked in the oropharynx and hypopharynx [64]. In the Dutch cohort of 35 patients studied with ambulatory polysomnography, the prevalence of OSAS was 46% (54% in children, 41% in adults) [64]. In the Norwegian national study, all individuals aged 5 years or older had sleep-disordered breathing and 18 of 19 met criteria for OSAS; reported snoring did not predict OSAS, and phenotypic severity was not associated with OSAS severity [65]. The three studies reviewed by Plomp et al. gave prevalences of 25%, 95% and 46%, with severe OSAS in 25%–41% and no evidence of a decrease with age [19]. A 2025 narrative review estimated the prevalence at about 70% and described worsening with age associated with the severity of mandibular hypoplasia [66]. The difference between studies reflects diagnostic criteria, ages and methods (ambulatory or laboratory). In 15 Norwegian adolescents and adults, obstructive sleep was present in 87%, and wake time after sleep onset, snoring and mean saturation were associated with worse health-related quality of life [67].

3.6. Feeding, oral cavity, salivary glands and teeth

Micro/retrognathia affects the temporomandibular joints (TMJ) and the masticatory muscles; the cleft palate is typically U-shaped, in the context of Robin sequence, and chewing and swallowing problems are frequent [1]. The review by Plomp et al. collated reduced mouth opening in 63%, malocclusion in 94%, a narrow hypopharynx in 84%, reported feeding difficulties in 68% and dry oral mucosa in 42%; cleft palate occurs in about 23% [19]. In the Norwegian series of 21 patients, ultrasound showed pathological salivary glands in about 48% (dysplasia 29%, aplasia 19%), and almost all had low salivary secretion, with no correlation with reported xerostomia [68]. In 19 individuals aged 5 to 74 years, all had orofacial dysfunction in at least two domains, correlated with severity, and a high orthodontic need [69]. At the Hospital for Rehabilitation of Craniofacial Anomalies of the University of São Paulo (HRAC-USP, Bauru), 60% of 15 orthodontically untreated patients had dental anomalies (agenesis 33.3%, mainly of the lower second premolars; enamel opacities 20%; ectopic eruption of the upper first molars 13.3%), and 66.7% had associated oral malformations [70]. The malocclusion is Angle class II with anterior open bite [1]. In the same Brazilian centre, bite force was 108.87 N in children with TCS versus 343.97 N in controls and 233.28 N in adults versus 412.07 N, and chewing ability was impaired in children but not in adults [71].

3.7. Nose, skull and other craniofacial structures

The nose shows a dorsal hump (73%), external deviation (up to 55%), a bifid or bulbous tip (55%) and septal deviation (82%); choanal atresia was described in 11% in one study [19]. Craniosynostosis is not part of TCS, but the skull may show brachycephaly with bitemporal narrowing; hypertelorism, a high-arched palate and macrostomia are also described [1]. Orbitozygomatic hypoplasia correlates with the severity of the downslanting fissures (r=0.461), ectropion, ptosis and eyelash abnormalities [63].

3.8. Extracraniofacial manifestations

In the cohort of 248 patients from four craniofacial units, 61 (25.4%) had at least one extracraniofacial anomaly: vertebral in 32, cardiac in 13, reproductive tract in 11, central nervous system in 7 and limbs in 7; the frequency tended to increase with severity [56]. Spinal anomalies affect about 10% (scoliosis, spina bifida occulta), and pectus excavatum or carinatum and flat feet are described [1]. In the French series, 7 of 92 patients with TCOF1 (8%) had congenital heart disease; the defects described are atrial septal defect (ASD), ventricular septal defect (VSD), patent ductus arteriosus (PDA) and patent foramen ovale, and rarely hypertrophic cardiomyopathy [1, 12]. Pyloric stenosis, oesophageal atresia and one case of chronic intestinal pseudo-obstruction with an enlarged myenteric plexus have been described [1, 72]. Renal anomalies are rare (congenital bilateral hydronephrosis, duplicated collecting system); the Oxford Desk Reference states that they are not found in TCS, but the cohort of Beaumont et al. described them [1, 45, 56]. Limb anomalies (syndactyly of the toes, hypoplastic thumbs, carpal fusion) are rare and should prompt reconsideration of the diagnosis [1].

3.9. Development, speech and cognition

Speech is of poor quality in about 60% of children, because of the hearing loss and the structural defects [33]. Speech errors are associated with malocclusion (about 60%), velopharyngeal incompetence (30%) and hearing loss (50%), with hypernasality, hyponasality or mixed resonance [19]. The European consensus found no direct causal association between the genetic basis and cognitive deficits, but recognises a risk of acquired deficits due to visual and hearing impairment, speech problems, episodic hypoxia, sleep fragmentation and reflux [20]. Intellectual disability without microcephaly is uncommon and occurs mainly after neonatal airway compromise or with microdeletions encompassing TCOF1 and neighbouring genes [12, 73]. In the Hungarian series, a patient with a de novo truncating TCOF1 variant and a mild craniofacial phenotype had developmental delay, moderate intellectual disability and absent speech [40].

3.10. Gene-specific features

The craniofacial phenotype of POLR1B-related TCS4 is similar to that of the other genes, with bilateral malar and mandibular hypoplasia, microtia, coloboma and downslanting fissures, although few patients have been described [13]. Of the 6 patients of Sanchez et al., all had downslanting fissures and malar hypoplasia, 3 had cleft palate and 2 choanal atresia; in the neonatal period, 5 required nasogastric tube or gastrostomy feeding and 3 intubation or tracheostomy [9]. Of the 7 patients with TCS2 in the French series, all had mandibular hypoplasia, 6 downslanting fissures and/or malar hypoplasia and 5 conductive deafness; 1 in 6 had cleft palate and 1 in 6 a spinal malformation, and none had the rare anomalies described in TCS1, which may reflect the sample size [7]. The AR forms due to POLR1C or POLR1D have been described with variable severity, including mild phenotypes in sisters compound heterozygous for POLR1C [28, 49]. Heterozygotes for POLR1D c.163C>G have no findings [1]. In the South African cohort, the homozygous POLR1D variant c.261del caused TCS, and a Mexican family with POLR1D c.294_295del showed AD inheritance over three generations [11, 41].

3.11. Natural history and treated disease

The severity of the deformities does not increase with age, according to Marszałek-Kruk et al., but OSAS may worsen with growth and recur after surgery [20, 33, 66]. In the US children's hospitals neonatal database (2010–2023), the 103 infants with TCS admitted to a neonatal intensive care unit (NICU) had more tracheostomies (54% versus 11%) and gastrostomies (67% versus 35%) and fewer mandibular distractions (9.7% versus 28.2%) than 4210 infants with micrognathia of other causes, with in-hospital mortality of 1.9% versus 7.2% and a mean length of stay of 47.5 days [74]. These figures describe patients treated in tertiary centres and not the natural history. With appropriate airway management, life expectancy approaches that of the general population [1].

4. Clinical Recognition and Warning Signs

4.1. Prenatal period

Prenatal suspicion arises in the presence of micrognathia, ear anomalies (small or hyperechogenic pinnae), cleft palate, absent or hypoplastic nasal bone, suspected glossoptosis, abnormal fetal swallowing and polyhydramnios, the latter in severe cases [1, 2, 4, 75]. Very severe micrognathia may be visible from 12–14 weeks [23]. Mildly affected fetuses may go undetected: in one case in the series of Li et al., the 20- and 30-week scans were normal and the newborn required a tracheostomy [1, 4]. External auditory canal atresia is rarely detected in utero [4]. Prenatal detection should lead to referral to a specialised obstetric unit, because of the respiratory risk at delivery, and of the parents to a genetics clinic and multidisciplinary team [20]. There is no biochemical or genetic newborn screening for TCS; universal newborn hearing screening may identify the hearing loss [1].

4.2. Newborn and infant

The newborn with symmetrical malar and mandibular hypoplasia, downslanting palpebral fissures, coloboma or notching of the lower eyelid and bilateral microtia has a recognisable phenotype [1]. The emergency signs are respiratory distress due to mandibular hypoplasia and glossoptosis, sometimes with associated laryngomalacia, apnoea, cyanosis during feeding and choking episodes [47, 76]. During examination of a child with respiratory compromise, the position of the head and neck should not be changed, and oximetry should be continuous [23]. In the NICU, more than half of infants with TCS require a tracheostomy and two-thirds a gastrostomy [74]. When the diagnosis has not been made before delivery, neonatal death from respiratory compromise may occur shortly after birth [4].

4.3. Child, adolescent and adult

In the older child and the adult, suspicion is based on the combination of bilateral conductive hearing loss with malformation of the pinna and canal, malar hypoplasia, downslanting fissures and absent eyelashes in the lateral third of the lower eyelid [1, 45]. Snoring, breathing pauses during sleep, daytime sleepiness and poor weight gain suggest OSAS, which should be sought even in the absence of complaints, because reported snoring does not predict it [20, 65]. Many mildly affected adults are identified only when a more severely affected child is born; the parents of a proband should be examined carefully for subtle dysmorphic features, including zygomatic hypoplasia visible only on radiography [1, 4, 13].

4.4. Signs pointing to another diagnosis

Microcephaly, intellectual disability, limb anomalies, marked facial asymmetry, oesophageal atresia or renal anomalies should lead to consideration of another disorder or a contiguous deletion: mandibulofacial dysostosis with microcephaly (EFTUD2), 5q32 deletion with CAMK2A, Nager or Miller syndromes, or craniofacial microsomia [1, 11, 12]. Isolated congenital sensorineural hearing loss, with pigmentation abnormalities, points towards other groups of syndromic deafness, such as the Waardenburg syndromes [62]. The triad of micrognathia, cleft palate and airway obstruction without malar hypoplasia or coloboma corresponds to Robin sequence [23].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Clinical criteria

No consensus clinical diagnostic criteria have been published [1]. The clinical diagnosis rests on bilateral, symmetrical anomalies: downslanting palpebral fissures, zygomatic and mandibular hypoplasia and conductive hearing loss [1]. Suggestive findings are coloboma or notching of the lower eyelid, sparse or absent lower eyelashes and lacrimal canaliculi, malar hypoplasia due to hypoplasia of the zygomatic arch and lateral orbital rims, mandibular hypoplasia, preauricular hair displacement and conductive hearing loss due to ankylosis, hypoplasia or absence of the ossicles [1]. Abnormal limbs, microcephaly and intellectual disability are rare in TCS and should lead to consideration of other disorders [1]. The diagnosis is confirmed by identification of a heterozygous pathogenic or likely pathogenic variant in TCOF1, POLR1D or POLR1B, of biallelic variants in POLR1C or, rarely, POLR1D, or of a heterozygous 5q32-q33.1 deletion including TCOF1; a VUS neither confirms nor excludes the diagnosis [1].

5.2. Imaging

Hypoplasia or aplasia (discontinuity) of the zygomatic arch is seen on the occipitomental radiograph (Waters view), and retrognathia on the orthopantomogram; before molecular testing, clinical and radiological examination was sufficient for the diagnosis even in the least affected [1, 2]. When conductive hearing loss is identified before 6 months of age, GeneReviews recommends craniofacial CT with axial and coronal slices to assess the anatomy of the head and neck, the canal, and the middle and inner ear [1]. The European consensus recommends three-dimensional analysis (3D photogrammetry, MRI or CT) in the craniofacial assessment, 3D CT of the zygomatic arch and TMJ when mouth opening is limited or reconstruction is planned, and CT of the temporal bones before ear surgery; Plomp et al. recommend CT to assess middle ear malformations and the thickness of the temporal bone before a bone-anchored implant [19, 20].

5.3. Genetic testing

The European consensus recommends referring every patient with suspected TCS to a clinical geneticist and choosing the test according to the phenotype and family history, with consent: a targeted NGS panel with TCOF1, POLR1B, POLR1C, POLR1D, SF3B4 and DHODH, including an in silico panel on an exome or genome platform, or RNA study in selected cases, when the phenotype is suggestive; exome or genome when the phenotype is atypical or there was no confirmation with older analyses; multigene panel, exome or genome when other features are present [20]. GeneReviews proposes two strategies: serial testing, starting with sequencing of TCOF1, followed by POLR1B, POLR1C and POLR1D, and then targeted deletion/duplication analysis of TCOF1 and POLR1D; or, alternatively, a multigene panel including the four genes and the genes of the differential diagnosis, which maximises yield and limits VUS and findings unrelated to the phenotype [1]. CMA should be considered if TCS is accompanied by intellectual disability, to detect deletions including TCOF1 and neighbouring genes [1, 30]. In the French series, the strategy included TCOF1, POLR1D, POLR1C and EFTUD2, and the atypical patients with intellectual disability or microcephaly had EFTUD2 variants or 5q32 deletions [12].

Table source: [1]. These values are proportions of published variants detectable by each method, calculated from the Human Gene Mutation Database (HGMD), and not the clinical sensitivity of the test; the series of Bowman et al., with 5% large TCOF1 deletions, shows that deletions may be underestimated [1, 50].

5.4. Diagnostic yield

The Brazilian cohort found a TCOF1 variant in 26 of 28 families (93%) [26]. In a series of 182 patients with a clinical diagnosis consistent with TCS, sequencing identified a TCOF1 variant in 50.5% and an intragenic deletion in 5.2% [6, 50]. Teber et al. identified TCOF1 variants in 28 of 36 patients (78%) with an unequivocal clinical diagnosis, and four clinically unaffected parents were heterozygous [6, 46]. In the French series of 146 patients, the molecular diagnosis was 63% for TCOF1, 6% for POLR1D and 0 for POLR1C; subsequent exome sequencing identified POLR1B in 2 of 146 patients (1.3%), and 6 (4%) with a typical phenotype remained without a molecular defect [12, 13]. In the South African cohort of 10 patients (2017–2021), a panel with TCOF1, POLR1C and POLR1D gave a diagnosis in 6 (60%), five of them with novel variants; the panel did not analyse CNVs or deep introns and did not include POLR1B [11]. In three Mexican families, exome sequencing with about 100× coverage gave a diagnosis in 10 of 11 patients (90.9%) [41].

5.5. Limitations of the tests

A negative result does not exclude TCS. Exome sequencing and panels do not detect structural variants in non-coding regulatory regions, deep introns or insertions of transposable elements [29, 40]. Targeted deletion/duplication analysis may not define the breakpoints of large deletions or detect deletions of adjacent genes [1]. CNV detection from exome data is indirect: in a Mexican family, a possible heterozygous deletion of exons 1–2 of POLR1C, inferred from read depth (normalised coverage of about 0.46), was not validated and would be insufficient in a gene with AR inheritance [41]. Sanger sequencing does not detect low-level parental mosaicism; high-depth NGS and analysis of other tissues, including semen, are required when mosaicism is sought [4]. In typical cases without a diagnosis, RNA sequencing, optical mapping and long-read sequencing may identify deep intronic variants [29]. Sanger validations should target low-quality regions, such as homopolymers and low-depth areas [11].

5.6. Interpretation and report content

Variants are classified according to the criteria of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP) [1]. The report should state the gene, the versioned transcript (preferably MANE Select: NM_001371623.1 for TCOF1, NM_015972.4 for POLR1D, NM_203290.4 for POLR1C, NM_019014.6 for POLR1B), the description according to Human Genome Variation Society (HGVS) nomenclature at cDNA and protein level, the zygosity and the class [10]. Because the historical TCOF1 nomenclature uses NM_001135243, the same variant may appear with different numbers in reports and articles, and comparison should be made using the genomic coordinate [10]. For POLR1D, interpretation of a heterozygous variant requires weighing zygosity, variant type, segregation, population frequency and phenotypic concordance, because the same pathway produces AD and AR forms [11]. Examples of application of the criteria: TCOF1 c.3183G>C, the last nucleotide of an exon, with a donor gain predicted by SpliceAI of 0.89, was classified as likely pathogenic (PVS1_Supporting, PM6, PM2, PP3); de novo TCOF1 c.1371_1372insT as pathogenic (PVS1, PM2, PS2); POLR1D c.295G>C, inherited from an asymptomatic mother, as VUS (PM2, PP3) [11, 40]. Segregation analysis in the parents is essential to distinguish de novo variants, variants inherited from a mildly affected parent or with reduced penetrance, and mosaicism [1, 4].

5.7. Facial analysis tools

The Face2Gene deep learning algorithm, applied to 127 public images of TCS (81 of white individuals, 23 Chinese and 23 Indian), achieved a diagnostic accuracy of 93.7%, with lower confidence in Asian individuals and a lower prevalence of malar hypoplasia in Indian individuals; it is an adjunctive tool [77].

5.8. Prenatal and preimplantation diagnosis

With the familial variant known, targeted prenatal testing on chorionic villi or amniotic fluid and preimplantation genetic testing (PGT-M) are possible, but the variant does not predict the malformations or the severity, and reduced penetrance should be considered, particularly with the recurrent exon 24 deletion [1, 4]. Cell-free fetal DNA is not diagnostic, and any decision requires confirmatory invasive testing [4]. Without a molecular diagnosis in the family, prenatal diagnosis and PGT are impossible and surveillance relies on detailed ultrasound scans [20]. 3D ultrasound improves detection of mandibular hypoplasia and ear malformations and helps distinguish TCS from Robin sequence and other acrofacial dysostoses, as in the Brazilian case diagnosed at 33 weeks by Pereira et al., confirmed after birth by CT and pathology [4, 78]. Quantitative assessment of micrognathia (inferior facial angle) is not uniform across centres [4]. In a Chinese fetus of non-consanguineous parents, the scan at 22+1 weeks showed micro-retrognathia with an inferior facial angle of about 45° and "C"-shaped pinnae; fetal MRI at 24+1 weeks confirmed micro-retrognathia, glossoptosis and a narrow pharynx, with external auditory canals not visualised and a small middle ear cavity; trio exome sequencing on amniotic fluid, at 26+5 weeks, identified the de novo variant POLR1D NM_015972.4:c.294_295del p.(Gly99ProfsTer3), the same normalised description as in the Mexican family with AD inheritance; the family opted for termination, and study of the paternal semen showed no mosaicism [41, 79]. In the table of 15 cases with prenatal data compiled by Li et al. (2007–2025; China, USA, Greece and Taiwan), there was termination of pregnancy in 6 (40%), and the findings included polyhydramnios, micrognathia, microtia, hyperechogenic pinnae, cleft palate, absent nasal bone and suspected glossoptosis; a 5q32 microdeletion was identified by microarray at 28 weeks [4]. Ultrasound suspicion should lead to targeted molecular testing, preferably as a trio [4].

5.9. Practical algorithm

6. Differential Diagnosis

6.1. Comparative table

Table sources: [1, 15, 23, 30, 45, 62, 73, 80–86].

6.2. Mandibulofacial dysostosis with microcephaly

MFDM is the main molecular differential diagnosis in cases with intellectual disability. It combines malar and mandibular hypoplasia, congenital or postnatal microcephaly, mild to severe intellectual disability, external ear malformations and typically conductive hearing loss; 25%–35% have heart disease, thumb anomalies, oesophageal atresia, short stature, vertebral anomalies or epilepsy [73]. Smith's describes conductive hearing loss in 50%, sensorineural in 10% and mixed in 25%, delayed myelination in half of patients, walking and first word at a mean of 26 months and acquisition of speech in only 50% [80]. Oesophageal atresia affects 27% according to the Oxford Desk Reference and about 34% in the GeneReviews table [1, 45]. Unlike TCS, the palpebral fissures are not consistently downslanting, and the cleft palate is median; microcephaly may show catch-up in adulthood, which requires attention to the head circumference recorded in childhood [45, 73]. At least 126 cases with an EFTUD2 variant have been published; 75% are de novo, 19% inherited from an affected parent, and germline mosaicism has been documented in 6% [73, 80]. In TCS, cardiac and oesophageal malformations are not part of the classical picture, and head circumference is usually normal [73].

6.3. Nager and Miller acrofacial dysostoses

Nager syndrome has the face closest to TCS: malar hypoplasia in 97%, downslanting fissures in 92%, micrognathia in 89%, partial absence of the lower eyelashes in 81%, cleft palate in 65%, ear anomaly in 93% and conductive deafness, usually bilateral, in 88% [82]. What distinguishes it are the upper limb defects, present in all patients, typically preaxial [82, 87]. Neonatal severity is greater than in TCS: in the US neonatal database, the 11 infants with Nager had median Apgar scores of 2 and 6, intubation at birth in 91%, tracheostomy in 72.7% and mortality of 27.3%, versus 1.9% in TCS [74]. Rodríguez acrofacial dysostosis represents the severe end of the spectrum, with fibular aplasia in lethal cases [82, 83]. SF3B4, a spliceosome component, accounts for about 60% of cases; birth prevalence in Finland was estimated at 0.3/100,000 [45, 85].

Miller syndrome combines a TCS-type face with lower eyelid coloboma and postaxial deficiency of all four limbs, with absence of the 5th digit and ulnar hypoplasia; intelligence is usually normal, hearing impairment is present in about 20%, and ectropion and facial asymmetry worsen with age [45, 81]. It is AR, due to biallelic variants in DHODH, the gene for dihydroorotate dehydrogenase in de novo pyrimidine synthesis, and was the first example of a rare Mendelian disorder whose gene was identified by exome sequencing [45, 81]. Dihydroorotate accumulates in plasma and urine and can be measured in dried blood spots by liquid chromatography–tandem mass spectrometry, which allows biochemical screening; in a 4-year-old boy with a plasma dihydroorotate of 3.5 µmol/L (reference <0.5), oral uridine (100 mg/kg/day) for 3 months reduced plasma dihydroorotate without improving behaviour, a use not authorised for this indication [88]. Neonatal cholestasis may progress to persistent liver disease [81]. About 30 to 40 cases have been described [20, 86].

6.4. Other ribosomopathies and the Pol I pathway

Cincinnati-type acrofacial dysostosis, due to heterozygous POLR1A variants, shares mandibulofacial dysostosis, micrognathia, eyelid coloboma, clefts and microtia, but is associated with neurodevelopmental abnormalities, heart disease and limb defects with bowed femurs [1, 32, 89]. The zebrafish polr1a model reproduces the TCS phenotype [13]. A 2025 report with the POLR1A variant NM_015425.6:c.1488G>T p.(Met496Ile) and overlap with several craniofacial syndromes proposed a spectrum of Pol I-related disorders [90]. Diamond-Blackfan anaemia with mandibulofacial dysostosis (RPS28, TSR2) is distinguished by macrocytic anaemia; about one-third of patients with DBA have craniofacial anomalies [1, 73]. A transcriptomic analysis in zebrafish found 17 differentially expressed genes shared between the polr1c and sf3b4 mutants, mostly in the same direction, suggesting common downstream pathways between TCS and Nager syndrome [91].

6.5. Craniofacial microsomia and Robin sequence

Craniofacial microsomia (oculo-auriculo-vertebral spectrum) is mostly unilateral and, when bilateral, asymmetrical; the coloboma, when present, affects the upper eyelid, and epibulbar dermoid and vertebral anomalies are suggestive [23, 45, 92]. Facial asymmetry is present in about 65% of patients with microsomia and in 58% of those with MFDM [73]. Symmetry and coloboma of the lateral third of the lower eyelid point to TCS [92]. FOXI3 variants accounted for 3.1% of 670 individuals with microsomia, and SF3B2 for about 3% of cases without a family history and 25% of familial cases [1].

Robin sequence is defined by the triad of micrognathia, U-shaped cleft palate and upper airway obstruction [23]. In the series of 74 patients of van den Elzen et al., two-thirds were isolated, and among the syndromic cases Stickler syndrome accounted for 34% and 22q11.2 deletion for 11%; in another series, fewer than 20% were isolated [23]. In deformation, mandibular growth normalises at around 2 years of age, unlike syndromes with an intrinsic mandibular anomaly, such as TCS [1, 23]. Some isolated forms result from disruption of a long-range cis-regulatory element with dysregulation of SOX9 [1]. Singh and Bartlett identified TCS in 52 of 266 individuals with congenital mandibular hypoplasia, without molecular confirmation [93]. 3D ultrasound helps distinguish the two conditions in the prenatal period [4]. The Brazilian book Síndromes Genéticas: Reconhecer para Tratar includes TCS in the differential diagnosis of Pierre Robin sequence, distinguishing it by the antimongoloid slant of the palpebral fissures, lower eyelid coloboma, hypoplasia of the zygomatic arches and microtia [94]. The Oxford Desk Reference also includes in the differential diagnosis of micrognathia SATB2-associated syndrome (severe neurodevelopmental impairment and cleft palate in about 50%), Potter sequence and oromandibular-limb hypogenesis (hypoglossia, transverse limb reduction, sometimes with Moebius sequence) [23].

6.6. Auditory differential diagnosis

In a child with an ear anomaly and hearing loss, looking for lower eyelid coloboma (TCS), epibulbar dermoid and upper eyelid coloboma (microsomia), iris or retinal coloboma (CHARGE), branchial pits or fistulae (branchio-oto-renal syndrome), head circumference (MFDM) and the limbs (radial ray in Nager, ulnar ray in Miller, triphalangeal thumbs in Townes-Brocks) guides the diagnosis [45]. Eyelid defects also occur in rare facial cleft syndromes and, occasionally, in amniotic band sequence [95]. Malformation of the tympanic membrane almost always produces conductive hearing loss; a sensorineural component should suggest other syndromes [45, 62]. Pigmentary changes, such as a white forelock or piebaldism, with sensorineural deafness, point to the Waardenburg syndromes, which account for more than 2% of childhood deafness [62]. Renal ultrasound is indicated in a child with an ear anomaly, unless there is a clear diagnosis of TCS according to the Oxford Desk Reference; GeneReviews, which describes rare renal malformations, recommends it in the initial assessment [1, 45].

7. Treatments and Risk Management Strategies

7.1. Sources of recommendations and organisation of care

GeneReviews states that no clinical practice guidelines have been published for TCS and bases its recommendations on the authors' experience [1]. The two reference documents are the ERN CRANIO European clinical consensus on facial dysostosis syndromes (Van Roey, Irvine et al., 2024), obtained by the Delphi method with 92 experts from 13 centres in 9 countries and 3 patient representatives, based on a literature search from January 1985 to February 2023, and the systematic review by Plomp et al. (2016), with 63 included studies (search from 1985 to January 2014) and recommendations graded according to the American Society of Plastic Surgeons scale (ASPS: levels of evidence I–V, grades of recommendation A–E) [19, 20]. A consensus sits one level of evidence below a clinical practice guideline [22]. In the consensus, agreement was deemed reached when the mean on the 1-to-9 Likert scale was ≥7.00, with at most one outlier and at least five votes; over the three rounds, 92 of 196, 58 of 98 and 19 of 27 consensus statements were obtained [20]. The consensus statement numbers are given in parentheses in the text. According to the consensus supplementary table, of the 168 consensus statements 118 have level of evidence V (expert opinion), 30 level IV, 19 level III and 1 level II (Oxford scale); for example, routine screening for OSAS (90) and screening for extracranial anomalies (182) are supported by level III, and the intervals for hearing surveillance (146, 214) by level V [20]. The American Cleft Palate-Craniofacial Association parameters (ACPA, 2018), generic for craniofacial differences, recommend an interdisciplinary team with a sufficient volume of patients, genetic and dysmorphological screening with referral for full genetic assessment, audiological assessment, instruction of carers on the signs of otological disease, hearing loss and sleep apnoea, otolaryngological, surgical or cardiopulmonary assessment in the infant at risk of respiratory obstruction, consideration of mandibular distraction in obstruction due to micrognathia, intensive care available for any procedure that may compromise the airway, and regular follow-up at least until adolescence [96].

Care is provided by a multidisciplinary team with expertise in craniofacial reconstruction, airway, eyes and lacrimal pathways, speech, feeding and swallowing, dental care and orthodontics, hearing, extracranial anomalies, psychology, cognition, genetics and coordination (2, 3); when national expertise is insufficient, ERN CRANIO is consulted (18) [20]. GeneReviews includes in the team the clinical geneticist, plastic surgeon, head and neck surgeon, otolaryngologist, oral surgeon, orthodontist, audiologist, speech and language therapist and psychologist, and organises the problems into three age groups graded by severity [1, 97, 98]. Parents and patients should receive information on the consequences of the disorder, the organisation of care and the individual plan before each intervention (5–9, 12), and contact with patient associations (16) [20]. Surgical treatment does not differ between the four genetic subtypes; priorities are defined by age, location and severity [33]. The FDS share the approach, with additional attention to Nager and Miller syndromes, which are more severe (23, 24) [20].

7.2. Airway in the neonatal period

When there is prenatal suspicion, GeneReviews recommends detailed fetal ultrasound, high-risk obstetric and neonatology consultation, and alerting the delivery team to the risk of airway compromise; in fetuses with predicted obstruction, an ex utero intrapartum treatment (EXIT) procedure may be planned, in which the head and neck are delivered by caesarean section for oral intubation or tracheostomy during birth [1]. Delivery should take place in a centre capable of managing the neonatal airway [4, 20]. Respiratory problems in the first year require a complete examination of the airway, and all levels should be assessed by a multidisciplinary team before any intervention (78, 79) [20]. Standard surgical interventions include restoring patency of the nostrils and choanae and mandibular distraction [1].

In the supplementary algorithm of Plomp et al., the newborn without acute respiratory distress is kept under clinical observation for 24 hours and undergoes polysomnography (level I, II or III); with no OSAS or mild OSAS, prone positioning is considered, as in Robin sequence; moderate to severe OSAS is managed with oxygen, high-flow nasal therapy, continuous positive airway pressure (CPAP) or bilevel positive airway pressure (BiPAP) or a nasopharyngeal tube; severe acute respiratory distress with endotracheal intubation, followed by airway endoscopy and, depending on the level of obstruction, tracheotomy or other surgery; respiratory distress should be reassessed every 3–6 months (grade D); at 9 months, in the child with cleft palate, polysomnography with simulated closure is performed, and repair is postponed if there is respiratory distress [19]. Nasopharyngeal tubes are temporary and are also used in newborns with severe glossoptosis [66].

7.3. Obstructive sleep apnoea syndrome

Laboratory polysomnography is mandatory in all newly referred patients, at any age (Plomp, grade B); the consensus considers it the reference standard (83) and recommends routine screening for OSAS in children and adults (90), without agreement on frequency (92), with high frequency when there are symptoms such as abnormal breathing during sleep or abnormal growth curves (91) [19, 20]. The Epworth scale in adults and the Brouillette score in children are not reliable for screening (85; Plomp, level II) [19, 20]. Overnight oximetry, lateral cephalometric radiography, airway MRI and video sleep studies in infants are screening alternatives when access to polysomnography is limited [66].

In severe OSAS, the level of obstruction is determined by flexible laryngoscopy in the otolaryngology clinic as a first approach (Plomp, grade C), and in children sleep endoscopy is useful (grade C); the consensus recommends airway endoscopy or drug-induced sleep endoscopy (DISE) to assess the dynamics (86, 88), routinely in tracheostomised patients and in severe OSAS (207), combined with other procedures to reduce the number of anaesthetics (208) [19, 20]. Surgery should be tailored to the level of obstruction, and single-level surgery does not resolve most cases, because the obstruction is multilevel [19, 64]. In moderate or severe multilevel OSAS, non-invasive ventilation (NIV; CPAP or BiPAP) or tracheostomy precede definitive interventions (94, 206); otolaryngological assessment should look for adenotonsillar hypertrophy (97) and treat it by adenotonsillectomy (98); if this fails and the obstruction is due to micrognathia, mandibular distraction is considered (81); all levels should be treated (93) [20]. In severe OSAS, candidacy for distraction requires endoscopy, videolaryngoscopy and clinical examination (104) [20]. OSAS may recur after distraction, and screening should be maintained (95), with polysomnography at least 6 weeks after distraction (89); in tracheostomised patients, polysomnography is performed with the cannula capped (84) [20].

CPAP is first line in mild to moderate OSAS, but adherence is limited: in the general population, 8% discontinue after the first night and 50% in the first year [66]. Mandibular advancement devices serve less severe cases, and advancement genioplasty is a less invasive alternative in mild cases [66]. Hypoglossal nerve stimulation, with a success rate estimated at about 80% in the general population, was used for the first time in TCS in 2023; bilateral stimulation has been described for complete concentric velopharyngeal collapse, and the combination of bilateral stimulation with CPAP is emerging as an alternative to tracheostomy, with evidence limited to case reports [66].

7.4. Mandibular distraction and tracheostomy

Mandibular distraction osteogenesis (MDO) is most effective when there is a vital risk in the first year from nocturnal apnoea due to mandibular hypoplasia and glossoptosis [33]. Plomp et al. consider it safe for the airway between the 1st and 4th year and describe, when it fails, distraction genioplasty with hyoid advancement (n=5); decannulation is not always possible after distraction, because of the multilevel obstruction [19]. The consensus recommends evaluating MDO results separately from those of Robin sequence (39), because success and complications are worse in FDS during growth, and informing parents of all the consequences (44) [20]. In the US neonatal database, distraction was used in 9.7% of infants with TCS and tracheostomy in 54% [74]. An Australian report of early distraction for established obstruction, reviewed at skeletal maturity and reassessed more than 30 years later, documents very long-term follow-up, but the abstract does not detail the results [99]. Distraction and bilateral sagittal split osteotomy reduce the apnoea–hypopnoea index (AHI), with a risk of overcorrection extrapolated from Robin sequence [66].

Tracheotomy is sometimes unavoidable in severe OSAS (Plomp, grade C) and was required in up to 41% of patients in the series reviewed [19]. The duration is indeterminate (212), and the family should be informed that decannulation is not always possible (210) [20]. Decannulation is considered when the cannula can be capped during the day, the in-hospital sleep study with the cannula capped and endoscopy show no significant obstruction, and the multidisciplinary meeting has discussed swallowing safety, the sleep and endoscopy results, ease of intubation and the family's preference (108a, 108b); the criteria vary between centres, partly because of the distance to hospital [20]. Cleft palate may reduce the probability of tracheotomy, according to a study across several syndromes [19].

7.5. Anaesthesia and the perioperative period

The narrow airway makes intubation difficult, and assessment of the level of obstruction is essential [2]. Before each anaesthetic induction there should be an airway plan defined by the anaesthetic team (99), with competent staff and equipment, techniques beyond direct laryngoscopy (101), a laryngeal mask when intubation is not needed (102) and a "difficult intubation" card (103) [20, 100]. After palatoplasty, a high-dependency bed should be available to monitor breathing (70) [20]. Grouping procedures under a single anaesthetic reduces the number of inductions (208) [20].

7.6. Feeding, swallowing and growth

A nasogastric tube or gastrostomy is used as needed, protecting the airway [1]. The consensus recognises a risk of aspiration with fatal or near-fatal outcomes and recommends routine screening for feeding problems (174), assessment by a speech and language therapist if suspected (175) and early feeding training, particularly with gastrostomy or tracheostomy (141) [20]. Length and weight are measured annually between 1 and 6 years (177), and inadequate growth should prompt a search for OSAS and feeding problems (178) [20]. Plomp et al. consider regular follow-up of height and weight mandatory and, if there are difficulties, monitoring of feeding with a speech and language therapist and dietitian (grade C) [19].

7.7. Cleft palate, cleft lip and speech

Cleft lip is treated according to cleft lip and palate guidelines (68), with lip repair at around 6 months and macrostomia repair before 1 year [20, 33]. GeneReviews places cleft palate repair at 1–2 years, and in its summary at around 1 year [1, 101]. The consensus allows closure if the obstructive AHI is ≤5 without significant CO2 retention, or if polysomnography is normal with or without a palatal plate (69), and recommends informing the family that a respiratory problem may arise after surgery (71); there is no evidence on the technique or on the optimal timing (72a, 72b) [20]. Plomp et al. found no evidence that timing should differ from that for non-syndromic clefts, report a higher risk of palatal fistula and recommend polysomnography with simulated closure before surgery, postponing it if there are respiratory problems [19].

Pre-speech and babbling are assessed at diagnosis (213), with screening at 2 years (135) and follow-up by a speech and language therapist as for non-syndromic clefts (136); therapy is delivered close to home under the supervision of the centre (139) and may be needed in adulthood (140) [20]. Plomp et al. consider a speech and language therapist mandatory from 2 years of age (grade C) [19]. Velopharyngeal insufficiency is assessed by videofluoroscopy and nasendoscopy (138) and corrected with a double buccal myomucosal flap (143) or other techniques, weighing the risk of OSAS (142) [20]. Speech therapy should start before and continue after structural normalisation, with attention to tracheostomised children and use of gestures and vocalisations [33].

7.8. Hearing

Audiological screening should be performed soon after birth (146), and otological and audiological assessment at least every 3 years, regardless of the severity of the malformations (214), because minor malformations may be accompanied by severe hearing loss and require early aiding (162) [20]. Plomp et al. recommend otolaryngological screening with otoscopy and audiometry from 3 months of age (grade B), with softband aids if there is hearing loss, and repeat otolaryngological screening at 4 years, the age at which their algorithm places the bone-anchored aid; GeneReviews recommends audiological assessment annually or as needed [1, 19]. The ACPA parameters, common to craniofacial differences, call for a complete diagnostic assessment by 3 months if newborn screening is failed, behavioural assessment from 9 months repeated at least every 6–12 months until 5 years and then annually until adolescence if hearing is normal, and conventional bone conduction amplification in microtia or atresia, with an osseointegrated implant as an option [96]. Air conduction aids are the first choice if the anatomy allows and the result is acceptable (215); with middle or external ear malformation and conductive or mixed hearing loss, bone conduction devices are used (151); implantable devices are reserved for when conventional ones (behind-the-ear, adhesive amplification, softband) are contraindicated or insufficient, or when reconstructive middle ear surgery is not feasible (152) [20]. The options include active middle ear implants and percutaneous or transcutaneous, passive or active bone conduction devices (144, 153); the choice is individualised (216), and the family should be informed about the principle, advantages, longevity, upgrades and MRI compatibility (145) [20]. Maximum power output (MPO) is used to compare devices (149), and audiometry is performed before and 4 weeks after acclimatisation (161, 220) [20].

Plomp et al. consider that implantation of a bone-anchored hearing aid (BAHA) significantly improves hearing (level II) and that it seems unavoidable, preceding auricular reconstruction (grade B); in the text of the review, the softband aid is used in the first years and the BAHA not before 3 years of age, whereas the table of recommendations indicates "from 4 years" [19]. The consensus diverges on this point: it favours air conduction when possible and sets no age for implantation [20]. There is no conclusive evidence of benefit from reconstructive surgery of the canal or middle ear (158), which is technically difficult (Plomp, level IV) and frequently disappointing; bone conduction devices are preferred (159) [19, 20]. GeneReviews reserves canal and middle ear reconstruction for bilateral microtia and narrow canals, after reconstruction of the pinna [1]. Hearing surgery plans and implants should be coordinated with the auricular reconstruction surgeon (204) [20].

Table sources: [19, 61, 102]. The systematic review by Marinac et al. gathered only 9 studies with extractable data, of level of evidence III–IV at most; the mean preoperative threshold was 55–70 dB and the mean improvement about 30 dB with bone conduction devices or conventional aids, and the authors issue no general recommendations owing to the scarcity of data [102]. An otolaryngologist with TCS and a bilateral Bonebridge implant described thin temporal bone, a low sinodural angle and fixation of the implant over the dura with a maxillofacial plate [102].

7.9. Eyes and eyelids

Ophthalmological assessment should take place at diagnosis, looking for colobomas and pseudocolobomas, lagophthalmos, lacrimal drainage anomalies, refractive errors and strabismus (127), and follow-up should monitor refractive errors, strabismus, blepharoptosis, epiphora, exposure keratopathy and amblyopia (128), with at least one assessment after craniofacial surgery (130) [20]. Plomp et al. recommend screening at around 1 year and regular follow-up (grade C), and Ng et al. perinatal assessment in all children [19, 59]. Lagophthalmos with a risk of exposure keratopathy is treated with protective ointment and an eye pad and, if necessary, surgery (62); exposure keratitis has medical treatment [1, 20]. Blepharoptosis or blepharophimosis with a risk to vision require prompt correction and orthoptic follow-up (63, 64); without functional consequence, there is no critical timing (65), and correction is preferably performed after periorbital craniofacial surgery, in a separate session (66) [20]. Mild epiphora is treated conservatively (132); severe epiphora by punctoplasty, balloon dacryoplasty, dacryocystorhinostomy or conjunctivodacryocystorhinostomy (133) [20].

The indication for correcting the eyelid deformities is imminent functional problems (corneal erosion, desiccation, incomplete closure); corrections are often multiple, and the family should be informed of the residual deformity (Plomp, grade B) [19]. Colobomas with a defect greater than one-third of the eyelid margin require timely surgery; in the algorithm of the same review, severe coloboma with corneal dryness in the newborn is treated with protective ointment and occlusion, followed by eyelid reconstruction within 3 months [19]. Colobomas are managed mainly with lubrication, but those threatening the cornea should be repaired in infancy [59]. For lower (pseudo)colobomas, the consensus describes autologous dermal fat grafting and subperiosteal malar elevation with pedicled muscle or myocutaneous flaps from the upper eyelid (59), but the evidence on the techniques is insufficient (60) and the anomalies frequently persist after surgery (61) [20]. Downslanting fissures are corrected from preschool age with an upper-to-lower eyelid myocutaneous flap and canthopexy [1, 33]. In the Philadelphia cohort, 40% of children underwent 86 periorbital procedures, and lateral canthopexy with malar fat grafting had fewer reoperations than other techniques (0% versus 72%; P=0.004) [59]. The same group proposed a visual loss risk scale (58% high risk, 18% moderate, 24% low), associated with acuity (P=0.050) [59].

7.10. Craniofacial reconstruction

Planning is carried out by the team (25), and the timing and sequence of interventions are individualised according to severity and function (mouth opening, speech, feeding, breathing during sleep, eyelid closure, hearing) (26) [20]. Bone reconstruction precedes soft tissue correction and may prevent progression of facial asymmetry [1]. The evidence is insufficient to define a standard surgical treatment, in particular for the orbitozygomatic and maxillary regions (Plomp; consensus, 55) [19, 20].

Zygomatic and orbital reconstruction is performed when cranio-orbito-zygomatic bone development is complete, at around 5–7 years according to GeneReviews, and not before 5 years, except in severe corneal exposure, according to Marszałek-Kruk et al. [1, 33]. The consensus recommends weighing the psychological benefit of orbitozygomatic reconstruction before maturity against restriction of maxillary growth and the need for later corrections (53); if it is bony, postponing it to 6–12 years reduces resorption (54) [20]. The techniques described are osteotomies, calvarial bone grafts (free or vascularised), costal cartilage, implants (mainly in adults) and fat transfer (Plomp, grade D); calvarial grafts resorb in almost all patients, and inorganic implants carry a risk of infection and displacement [19]. The orbital floor may be reconstructed with allogeneic or custom implants [33]. Orthognathic surgery may be combined with autologous fat transfer, including suborbital dermofat grafts to reduce ectropion (29, 30) [20].

Counterclockwise correction benefits maxillary or mandibular hypoplasia, improves projection and may open the posterior nasopharyngeal airway (27); counterclockwise distraction may give more palatal rotation (33); without a respiratory problem, definitive orthognathic surgery is performed at skeletal maturity (28) [20]. GeneReviews indicates orthognathic therapies typically before 16 years and stages maxillomandibular reconstruction by severity, according to the Kaban classification: type I (mild) and IIA (moderate) at 13–16 years; type IIB (moderate to severe) at skeletal maturity, at around 16 years; type III (severe) at 6–10 years [1, 33]. The indications for treating severe maxillary hypoplasia are malocclusion, obstruction at that level, profile and support of the lower eyelid; midface correction is performed by Le Fort I osteotomy (after 16 years) or Le Fort II with cranial grafts (Plomp, grade D) [19]. The indications for mandibular surgery are severe obstruction at the base of the tongue, malocclusion and severe feeding problems; in mild deformities without OSAS, genioplasty may suffice (grade D) [19]. Rhinoplasty, if needed, is performed after orthognathic surgery [1]. The classification systems (Hayashi, O.M.E.N.S., four-type orbitozygomatic classification) have no predictive value [19]. Assessment is made with 3D analysis (34, 35), and satisfaction with appearance should be measured with patient-reported outcome measures (PROM) (38) [20].

Refractory TMJ disease, such as ankylosis, may require total joint replacement (47) [20].

7.11. External ear and nose

Reconstruction of the pinna is performed after 6 years according to GeneReviews, and with autologous cartilage from 9 years according to the text of Plomp et al., whose supplementary algorithm places auricular and orbitozygomatic reconstruction (including the eyelids) between 8 and 10 years, septo(rhino)plasty at 17 years and mandibular and/or maxillary surgery and lacrimal surgery at 18 years, whenever indicated [1, 19]. Auricular reconstruction is hampered by the low hairline and by the unavailability of the temporal flap if it has already been used for the zygoma [19]. The consensus recommends following the principles of the European craniofacial microsomia guideline (58), choosing between synthetic implants and costal cartilage (198) on an individual basis, according to age, tissues, anatomy, extracranial anomalies and preferences (199), and using osseointegrated implants for an epithesis after failure (203) [20]. Significant septal deviation with obstruction is corrected by septoplasty after the pubertal growth spurt (Plomp, grade C), and septorhinoplasty is performed from 17 years [19].

7.12. Dental and orthodontic care

Dental assessment should be six-monthly (164), because xerostomia due to salivary disease and mouth breathing and a soft diet favour calculus, caries and periodontal disease; at least one clinical and radiological screening for dental anomalies should be performed (168), the team orthodontist monitors the dentition and jaws (165), and dental care is lifelong (167) [20, 69]. GeneReviews likewise recommends dental examination every 6 months and orthodontics as needed [1].

7.13. Extracranial anomalies and psychology

Extracranial anomalies should be screened for at diagnosis, with attention to vertebral and cardiac anomalies (182): spina bifida occulta, dysmorphic vertebrae, reduced intervertebral space, scoliosis, pectus (183) and ASD, VSD or PDA (184) [20]. GeneReviews recommends echocardiography, renal ultrasound and assessment of the spine and limbs in the initial assessment, and investigation for pyloric stenosis, oesophageal anomalies or pseudo-obstruction if feeding problems persist [1]. Treatment of cardiac, gastrointestinal, renal and limb anomalies follows usual practice [1].

Psychology consultations with the family on motivations, expectations and coping strategies should precede reconstruction (187), with psychological or educational support for the parents (193), cognitive assessment by a (neuro)psychologist if there are concerns (190), play therapy (195) and assessment of anxiety (191) and of psychological trauma after life-threatening situations (192); surgery combined with psychological support improves long-term functioning (188) [20]. Plomp et al. recommend creating realistic expectations about the results (grade B), a psychology consultation at around 4 years (grade C) and counselling of the parents on the child's assertiveness and coping (grade D) [19]. The psychological functioning of adults with rare facial clefts is broadly normal, but there is stigmatisation and avoidance, a lower frequency of having a partner or children and a tendency towards internalising problems; objective severity does not predict acceptance, and long-term satisfaction with the ears, profile and eyelids after surgery is lower [19].

7.14. Drug therapy and regulatory status

There is no specific drug treatment for TCS authorised by the US Food and Drug Administration (FDA) or the European Medicines Agency (EMA), and a search of ClinicalTrials.gov (10 October 2026) found no interventional trial with TCS as the stated condition [21]. The medicines used in TCS serve general indications (ocular lubricants, anaesthesia, perioperative analgesia) and have no specific dosing in the sources consulted. Botulinum toxin type A has been used in a single patient (n=1): a 21-year-old man with TCS and lower colobomas had had, for as long as he could remember, an inferolateral distortion of the right lower eyelid with retraction of the cheek skin on smiling, which led him to avoid smiling in public; 18 units of onabotulinumtoxinA (Botox®) were injected along the line of retracted skin, identified by asking the patient to smile, into the presumed anomalous muscle [24]. One week later the distortion had disappeared, which confirmed its muscular origin; it remained absent at 2 months and partially reappeared at 4 months; the patient opted for surgery under general anaesthesia, with resection, through a lower eyelid incision, of a malpositioned or supernumerary muscle, with no recurrence at 2 months postoperatively [24]. The authors propose the toxin as a diagnostic test that distinguishes a muscular anomaly from a ligamentous, fibrous or scar-related change, before definitive surgical correction [24]. This is off-label use, in a single case report, with no basis for a routine recommendation, and botulinum toxin does not treat the coloboma itself [1, 24]. Uridine belongs to the differential diagnosis and not to TCS: in a 4-year-old boy with Miller syndrome, oral uridine at 100 mg/kg/day for 3 months, as a supplement and with ethics committee approval, reduced plasma dihydroorotate from more than 3 µmol/L to 0.6 µmol/L without adverse effects or changes in routine biochemistry, but did not improve behaviour and did not reduce urinary excretion; this is off-label use in a single patient [88]. N-acetylcysteine and other antioxidants, as well as p53 inhibition, have only preclinical data, with no human studies and no defined dose for prevention in pregnancy [37, 38, 103].

Bone conduction hearing devices are the regulated products of greatest relevance in TCS. The US labels set minimum ages: 5 years for the paediatric BAHA (1999) and for the Ponto, whose placement is contraindicated below that age; 12 years for the Bonebridge and for the Osia in the 2019 clearance, whereas the 2026 Osia label (K260902) indicates 5 years or older; all are intended for conductive or mixed hearing loss with a mean bone conduction threshold of up to 45–65 dB HL depending on the processor, and for single-sided deafness [104]. Below these ages, non-implanted devices such as the softband are used [19, 20].

Table sources: [19, 20, 104, 105]. In the EU, medical devices do not go through the EMA: they are certified by notified bodies, and the public EUDAMED database (consulted on 10 October 2026) shows the risk class and marketing status registered by the manufacturer, but not the indications or ages; registration in EUDAMED is not yet exhaustive, so the absence of the Bonebridge does not mean absence from the European market [105].

7.15. Clinical intervention studies and their limitations

Table sources: [59, 61, 64, 65, 71, 74, 102]. There are no randomised trials in TCS; the recommendations are based on series from craniofacial centres and expert opinion [19, 31]. The studies are retrospective or cross-sectional, with small samples, different criteria (for example, ambulatory or laboratory polysomnography) and populations selected by tertiary centres, and they do not allow indirect comparison between techniques. The differences between TCS and other micrognathias in the US neonatal database reflect treatment decisions and severity, and not the relative efficacy of each intervention [74]. Hearing gains are measured as short-term thresholds, with no generic quality-of-life data [102].

7.16. Recent developments and research (October 2025 to October 2026)

Preclinical antecedents with therapeutic relevance are the prevention of craniofacial anomalies in the Tcof1+/− mouse by p53 inhibition and by maternal antioxidants [37, 38]. None of these approaches has been tested in humans [21]. The registered trials with related terms address craniofacial microsomia, fat grafts with adipose-derived regenerative cells (University of São Paulo, NCT01674439; Hospital Sírio-Libanês, NCT03806361) or 3D-printed titanium implants (NCT07654647), without TCS as a condition; the absence of results in a search does not demonstrate the non-existence of studies [21].

7.17. Recommendations of the reference sources: agreements and divergences

Table sources: [1, 19, 20]. The divergence most relevant to practice concerns the first choice of hearing rehabilitation and the interval for hearing surveillance: the consensus, more recent and with patient participation, favours air conduction and a minimum interval of 3 years, whereas GeneReviews maintains annual assessment [1, 20].

7.18. Agents to avoid, pregnancy, lactation and transition

No drugs are specifically contraindicated in TCS in the sources consulted; the risks arise from the airway, and any sedation or anaesthesia requires the airway plan described in 7.5 [20]. Sleepiness questionnaires do not replace polysomnography [19, 20].

Fertility is usually normal [40]. The literature on pregnancy in women with TCS is limited to reports: a pregnant woman with exome-confirmed TCS and mechanical heart valve prostheses (for rheumatic valve disease, on warfarin) was delivered by caesarean section at 35 weeks and 5 days with a multidisciplinary approach and a healthy newborn; the anticoagulation was due to the prostheses and not to TCS [107]. In affected women, the anaesthetic airway plan applies to delivery, and the fetus has a 50% risk of inheriting the variant, with unpredictable severity, which justifies detailed ultrasound and delivery in a centre capable of managing the neonatal airway [1, 20]. The sources consulted describe no data on lactation in women with TCS.

Transition to adult care includes orthognathic and nasal surgery at skeletal maturity, surveillance for OSAS, which persists in adults (41% OSAS in Dutch adults; obstructive sleep in 87% of Norwegian adolescents and adults), lifelong dental care, speech therapy when needed and reproductive counselling [1, 20, 64, 67]. The timeline of the European registry extends beyond 17 years [106].

8. Reproductive Options and Family Counselling

8.1. Modes of inheritance

TCS is autosomal dominant in most cases, due to variants in TCOF1 and, less frequently, POLR1B or POLR1D; it is autosomal recessive in a minority, due to biallelic variants in POLR1C or POLR1D [1, 13, 27, 28]. Distinguishing the AD from the AR form in POLR1D is critical for the recurrence risk, carrier testing and reproductive decisions [11]. The European consensus recommends genetic counselling for parents planning a further pregnancy, with information on the recurrence risk, prenatal diagnosis and PGT (110, 111) [20].

8.2. Autosomal dominant form

Between 55% and 61% of probands with AD TCS have a de novo variant, and about 40% have an affected parent [1, 12, 108]; the Oxford Desk Reference and other sources give 60%–70% de novo [1, 40, 45]. The family history may appear negative because of unrecognised mild expression, reduced penetrance or parental mosaicism [1]. In a simplex case with a known variant, the parents are tested; without an identified variant, the parents undergo audiological assessment and an occipitomental radiograph, which may reveal mild hypoplasia or aplasia of the zygomatic arch [1, 109]. If the variant is not detected in the parents' leucocytes, it may be de novo or result from parental germline or gonadosomatic mosaicism; the mosaic parent may be unaffected or mildly affected [1, 110].

8.3. Autosomal recessive form

The parents are presumed heterozygous and both should be tested [1]. If only one is a carrier, the second variant may be de novo in the proband or postzygotic in a mosaic parent; apparent homozygosity may result from an undetected exonic deletion or from uniparental isodisomy [1]. Carriers are asymptomatic; siblings of the proband have a 25% chance of being affected, 50% of being carriers and 25% of being neither affected nor carriers; the children of an AR-affected person are obligate carriers, and the parents' siblings have a 50% chance of being carriers [1]. Carrier testing requires prior identification of the familial variants [1]. AR forms have been described mainly in consanguineous families [28].

8.4. Parental mosaicism

Mosaicism has been documented in both mothers and fathers [1, 110]. In the 2026 aggregated review, 6 of 275 individuals (2.18%) had confirmed or strongly suspected parental mosaicism, probably underestimated owing to the limits of Sanger sequencing [4]. The case of an unaffected father with about 10.84% variant alleles in blood and 8.74% in semen, who transmitted severe TCS to the fetus, led the authors to recommend testing the father first, in blood and semen, and considering PGT-M when there is mosaicism [4]. In the family of Ferraro et al., the healthy father was a gonadosomatic mosaic carrier of an SVA insertion transmitted to two half-siblings with different mothers, with direct consequences for the reproductive choices of his other children [29]. In TCS4, a father with mild dysmorphism had about 16% variant alleles in blood and the variant present in saliva [13]. The consensus recommends considering somatic and germline mosaicism in counselling (124) [20].

8.5. Cascade testing and asymptomatic relatives

The consensus recommends targeted testing of relatives of confirmed patients even without typical facial features (125), because of intrafamilial variability and reduced penetrance [20]. Transmission of the exon 24 deletion through several asymptomatic female carriers in a Hungarian family, and from an asymptomatic mother in the clinical case in section 10, illustrates this need [40, 47]. Identifying an asymptomatic carrier does not allow prediction of severity in the offspring, because the same variant produces mild and severe phenotypes in the same family [1, 52]. Testing of asymptomatic minors takes place within the framework of genetic counselling and with the consent provided for in the consensus (112) [20].

8.6. Family planning, prenatal and preimplantation diagnosis

Risk determination should ideally take place before pregnancy, and counselling should be offered to young adults who are affected or at risk; when the molecular diagnosis has not been confirmed, DNA banking is recommended [1]. With the familial variant identified, prenatal testing and PGT-M are possible; the variant does not predict the malformations or the severity, and reduced penetrance, particularly of the recurrent exon 24 deletion, should be discussed [1, 4]. Without a molecular diagnosis, prenatal diagnosis and PGT are not possible and surveillance relies on detailed ultrasound (121, 122), which may not detect mild forms [1, 20]. In the compilation by Li et al., 6 of 15 pregnancies with prenatal data ended in termination, decisions influenced by the foreseeable severity and by social and cultural values; the authors frame them within the principles of reproductive autonomy, proportionality and justice [4]. Perspectives on prenatal testing vary and the decision is personal [1].

A 2026 qualitative study with 18 people with visible hereditary conditions, including TCS, identified four themes in reproductive decisions: coping and adaptation; stigma and a sense of responsibility in choosing between natural conception, in vitro fertilisation with PGT and termination; managing uncertainty through prenatal testing and assisted reproduction; and the need for impartial genetic counselling, psychosocial support and peer contact [111].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Prevalence and what each figure measures

The TCOF1-specific prevalence was estimated by GeneGuidance at 0.8–1.1/100,000, by calculation and not by direct measurement; all estimates tend to underestimate the true frequency, because mild cases and carriers with reduced penetrance go undiagnosed [31]. No sex or ethnic predilection has been described [40]. The prevalences of Nager and Miller syndromes are unknown, with about 100 and 30 to 40 published cases, respectively [20, 85, 86].

9.2. Data by population

In the USA, the two most common diagnoses among hospital admissions of children with TCS, after the diagnosis itself, were tracheostomy and OSAS, and the most frequent procedures were airway examinations [18]. The US children's hospitals neonatal database documents 103 infants with TCS in NICUs between 2010 and 2023 [74].

In Europe, the French series of 146 patients provides the most widely used proportions by gene (TCOF1 63%, POLR1D 6%, POLR1B 1.3%, no diagnosis 4%) [12, 13]. The Norwegian national cohort of individuals aged 5 years or older documented OSAS in 18 of 19, and the Dutch Rotterdam cohort 46% of 35 [64, 65]. The multicentre cohorts from Rotterdam, London, Boston and Toronto described the ocular (194 patients) and extracraniofacial (248 patients) anomalies [56, 57]. In Hungary, three of four variants in five patients were inherited from asymptomatic mothers [40]. The German series of Teber et al. did not find the recurrent exon 24 deletion [46].

In Brazil, the cohort of 28 families had a TCOF1 detection rate of 93%, with mutations of paternal and maternal origin and no effect of paternal age [26, 53]. HRAC-USP, in Bauru, described dental anomalies in 60% of 15 patients and masticatory function in 27 patients [70, 71]. A case of prenatal diagnosis by 3D ultrasound was published by a São Paulo group [78]. The sources consulted contain no epidemiological data from Portugal, which does not demonstrate that such data do not exist.

In South Africa, an NGS panel gave a diagnosis in 6 of 10 patients, and the recurrent exon 24 deletion was described for the first time in an African family; the authors highlight the under-representation of African genomes in reference databases, which hampers variant interpretation [11]. In Mexico, exome sequencing gave a diagnosis in 10 of 11 patients from three families, and Latin American data remain scarce [41]. In Asia, the review by Ulhaq et al. suggested greater severity in patients with TCOF1 variants; Face2Gene had lower confidence in Asian individuals and malar hypoplasia was less prevalent in Indian individuals; most of the prenatal cases compiled by Li et al. come from China [4, 44, 77].

9.3. Special populations

Newborns with a prenatal diagnosis should be born in a centre capable of managing the neonatal airway [4, 20]. Tracheostomised children need particular attention to speech development and feeding [20, 33]. Adults remain at risk of OSAS, xerostomia and dental needs, and satisfaction with surgical results and psychosocial integration are persistent issues [19, 67, 69]. Affected pregnant women need an anaesthetic airway plan and targeted fetal assessment [20, 107].

9.4. Referral networks and patient associations

ERN CRANIO, coordinated from Rotterdam, publishes on its Standards of care page the consensus on facial dysostoses and the European guidelines on craniofacial microsomia, Robin sequence and cleft lip and palate, and is developing a European FDS registry [22, 106]. The Dutch association Laposa took part in the consensus [20, 112]. In the United Kingdom, Changing Faces provides TCS-specific information and the Treacher Collins Family Support Group is a registered charity (no. 1006300) [113, 114]. In the USA, the Children's Craniofacial Association, FACES: The National Craniofacial Association, the Foundation for Faces of Children and NORD are resources listed by GeneReviews [1, 115]. The Oxford Desk Reference lists CLAPA (United Kingdom), ACPA, Contact a Family and the Pierre Robin Network [23]. The sources consulted did not identify Portuguese or Brazilian associations specific to TCS.

10. Illustrative Clinical Case

10.1. Presentation

Zhang et al. (2025) described a Chinese boy, the second child of consanguineous parents without dysmorphism, born by vaginal delivery at 40 weeks, without asphyxia, weighing 3.5 kg and measuring 50 cm [47]. The couple's first pregnancy had ended in sudden fetal death at 32 weeks, in a fetus with ear malformation and mandibular hypoplasia, without autopsy; the maternal grandmother had an ear deformity and hearing loss [47]. The newborn was admitted to neonatology for severe choking episodes during oral feeding [47]. On examination he had normal pigmentation, downslanting palpebral fissures, a depressed nasal bridge, marked malar hypoplasia, mandibular hypoplasia, right microtia, left external auditory canal atresia with a preserved pinna and grade I cleft palate; a continuous murmur in the left 2nd intercostal space led to echocardiography, which showed a patent ductus arteriosus [47]. Full blood count and biochemistry were normal, and the child was discharged at the parents' request with feeding guidance [47]. The pedigree (Figure 1), the facial phenotype with microtia and downslanting fissures (Figure 2) and the timeline of the case (Figure 3) are available on the publisher's website: Figure 1, Figure 2 and Figure 3 [47].

10.2. Course

At 1 month he weighed 4.2 kg and measured 54.5 cm [47]. At 6 months, the Gesell developmental schedules showed quotients of 41.5 for gross motor (moderate deficit), 37.7 for fine motor (severe), 56.6 for adaptive (mild), 50.9 for language and 50.9 for personal-social (moderate), delay being defined by a quotient below 75 [47]. Auditory brainstem evoked potentials showed thresholds of 65 dB nHL on the right and 55 dB nHL on the left, and the child started hearing aids and rehabilitation [47]. At 1 year and 8 months, the ductus arteriosus had closed, the motor, adaptive and personal-social quotients were 77.7–80.5 (borderline) and language remained delayed (55.1) [47].

10.3. Molecular diagnosis

Exome sequencing identified a heterozygous frameshift variant in exon 22 of TCOF1, described by the authors as NM_001135243.2:c.3559delG p.(Ala1187GlufsTer21), classified as likely pathogenic (PVS1 and PM2_Supporting) and absent from ClinVar, HGMD and gnomAD [47]. Normalisation by VariantValidator gives NM_001135243.2:c.3559del p.(Ala1187GlnfsTer21) and, in the MANE transcript, NM_001371623.1:c.3562del NP_001358552.1:p.(Ala1188GlnfsTer21) (NC_000005.10:g.150392749del); the difference in the article's protein description corresponds to a notation error [10]. Sanger sequencing confirmed the variant in the patient and in the asymptomatic mother, indicating incomplete penetrance; the father and maternal grandmother were not tested [47]. The chromatograms and the diagram of the treacle domains are in Figure 4 and Figure 5 [47]. The authors propose that the variant eliminates the C-terminal NLS and the S1199 phosphorylation site, a target of the ATM/ATR kinases required for nucleolar localisation, with an effect analogous to that of the exon 24 deletions [47].

10.4. Management and comment

The plan described is staged: palatoplasty after 2 years, autologous reconstruction of the microtia in childhood, orthognathic surgery in adolescence, psychosocial support and regular auditory-verbal assessments [47]. The report does not mention polysomnography or airway endoscopy, which the European consensus and the review by Plomp et al. recommend for all patients, particularly with choking and cleft palate, and which the consensus requires before palatal closure (AHI ≤5) [19, 20]. The developmental delay at 6 months is more marked than usual in TCS; the authors acknowledge that consanguinity may increase the probability of other undetected variants, and the partial recovery at 20 months is compatible with the role of hearing loss and feeding difficulties in the delay [20, 47]. The case illustrates variable expressivity (carrier mother without signs, grandmother with an ear phenotype and severely affected previous fetus), the importance of parental testing and the need for counselling about the 50% risk in a future pregnancy, with unpredictable severity [1, 47].

A second case, published by Zhu et al. (2025), illustrates TCS2: a Miao boy with the POLR1D variant NM_015972.4:c.220dup p.(His74ProfsTer8), inherited from an unaffected father, had severe neonatal obstruction due to the tongue base and laryngomalacia, required a tracheostomy, was decannulated at around 10 months and still had conductive hearing loss at 2 years and 3 months, with a favourable course [76].

11. Prognosis and Long-Term Follow-up

11.1. Prognosis

Mortality is concentrated in the neonatal period and results from airway obstruction [1]. In the aggregated review of 275 individuals with a TCOF1 variant, 12 (4.36%) died shortly after birth or had a termination of pregnancy, with survival of 95.64%, in a sample biased towards severe cases [4]. In US infants admitted to NICUs, in-hospital mortality was 1.9% [74]. Late death from aspiration also occurs: a Hungarian patient died at 2 years from acute aspiration and a tracheostomised Dutch boy died at 4 years after choking [29, 40]. With appropriate airway management, life expectancy approaches that of the general population [1]. Intelligence and fertility are usually normal [1, 40].

The functional prognosis depends on hearing, speech, vision and the airway. OSAS persists in adults and there is no evidence that it decreases with age [19, 64]. Vision is good in at least one eye in most patients, but amblyopia is more resistant to treatment [58, 59]. Hearing rehabilitation with bone conduction devices improves thresholds by about 30 dB [102]. Satisfaction with surgical results is lower than in the general population, and acceptance depends more on self-perception than on objective severity [19]. OSAS is associated with worse health-related quality of life in adolescents and adults [67].

11.2. Initial assessment after diagnosis

11.3. Surveillance

The two recommendations on hearing are compatible: the consensus 3-year interval is a minimum, which the annual GeneReviews assessment satisfies, and the choice depends on age, language acquisition and use of a hearing device [1, 20].

12. Summary and Final Considerations

Treacher Collins syndrome is a bilateral, symmetrical mandibulofacial dysostosis resulting from disturbed ribosome biogenesis in cranial neural crest cells during a restricted window of embryonic development [1, 35, 37]. Four genes account for about 96% of cases: TCOF1, in 60%–90%, with dominant inheritance and truncating variants acting through haploinsufficiency; POLR1D, in 6%–8%, dominant or, rarely, recessive; POLR1B, in 1%–2%, dominant; and POLR1C, in about 1%, recessive [1, 12, 13]. Different biallelic POLR1C variants, which impair Pol III and not Pol I, cause a hypomyelinating leukodystrophy without facial dysostosis [15, 17]. Birth prevalence lies between 1/80,000 in US inpatient data and 2.0/100,000 in the European Orphanet estimate, with probable underestimation of mild forms [5, 18].

The phenotype combines downslanting palpebral fissures, lower eyelid coloboma, malar and mandibular hypoplasia, microtia with canal atresia and conductive hearing loss, with usually normal intelligence; intellectual disability, microcephaly and limb anomalies point towards a contiguous 5q32 deletion, mandibulofacial dysostosis with microcephaly or the Nager and Miller acrofacial dysostoses [1, 30, 73]. Expressivity varies within the same family, penetrance is reduced for some variants and parental mosaicism, mainly paternal, has been documented in blood and semen, so the parents of a proband should be examined and tested with appropriate methods [1, 4, 40]. A negative test does not exclude the diagnosis, because deletions, deep intronic variants and retrotransposon insertions escape exome sequencing and conventional panels [29, 50].

Management is multidisciplinary and rests on the ERN CRANIO European consensus and the systematic review by Plomp et al., in the absence of formal guidelines and randomised trials [19, 20]. The airway is the priority from the prenatal period: OSAS is frequent at all ages, multilevel and often severe, polysomnography should be performed in all patients, mandibular distraction has worse results than in Robin sequence and tracheostomy is required in a significant proportion of infants [20, 64, 65, 74]. Early hearing rehabilitation with air or bone conduction aids, ophthalmological surveillance for corneal exposure and amblyopia, cleft palate closure conditioned by respiratory function, staged craniofacial reconstruction up to skeletal maturity, lifelong dental care and psychological support complete the plan [1, 19, 20]. There is no specific drug therapy; p53 inhibition and antioxidants remain at the preclinical stage [21, 37, 38]. With appropriate airway management, life expectancy approaches that of the general population, and outcomes depend mainly on hearing, speech, vision, sleep and psychosocial integration [1, 67].

13. References

1. Barbosa M, Jabs EW, Huston S. Treacher Collins Syndrome. 2004 Jul 20 [Updated 2024 Jun 20]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–. https://www.ncbi.nlm.nih.gov/books/NBK1532/ PMID: 20301704.
2. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 362–3 (Treacher Collins syndrome).
3. Unger S, Ferreira CR, Mortier GR, et al. Nosology of genetic skeletal disorders: 2023 revision. Am J Med Genet A. 2023;191:1164-209. doi:10.1002/ajmg.a.63132 PMID: 36779427.
4. Li C, Huang W, Gan Z, Ling Y, Qiu L, Xiao Y, et al. Integrating imaging and genomics in prenatal Treacher Collins syndrome: evidence for practice and policy. Orphanet J Rare Dis. 2026;21(1):99. doi:10.1186/s13023-025-04094-4 PMID: 41857598. PMCID: PMC13001374.
5. Orphanet. Treacher-Collins syndrome (ORPHA:861). https://www.orpha.net/en/disease/detail/861
6. Online Mendelian Inheritance in Man, OMIM®. MIM 154500: Treacher Collins syndrome 1; TCS1 [Internet]. Baltimore (MD): Johns Hopkins University; [accessed 10 Oct 2026]. Available at: https://omim.org/entry/154500
7. Online Mendelian Inheritance in Man, OMIM®. MIM 613717: Treacher Collins syndrome 2; TCS2 [Internet]. Baltimore (MD): Johns Hopkins University; [accessed 10 Oct 2026]. Available at: https://omim.org/entry/613717
8. Online Mendelian Inheritance in Man, OMIM®. MIM 248390: Treacher Collins syndrome 3; TCS3 [Internet]. Baltimore (MD): Johns Hopkins University; [accessed 10 Oct 2026]. Available at: https://omim.org/entry/248390
9. Online Mendelian Inheritance in Man, OMIM®. MIM 618939: Treacher-Collins syndrome 4; TCS4 [Internet]. Baltimore (MD): Johns Hopkins University; [accessed 10 Oct 2026]. Available at: https://omim.org/entry/618939
10. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-68. doi:10.1002/humu.23348 PMID: 28967166. Accessed 10 Oct 2026: https://variantvalidator.org
11. Nevondwe P, Mudau M, Seymour H, Kerr R, Lombard Z, Krause A, et al. Molecular Characterisation of Treacher Collins Syndrome in a South African Cohort: Novel Disease-Causing Variants in TCOF1 and POLR1D. Mol Genet Genomic Med. 2026;14(9):e70295. doi:10.1002/mgg3.70295 PMID: 42706666. PMCID: PMC13550757.
12. Vincent M, Geneviève D, Ostertag A, Marlin S, Lacombe D, Martin-Coignard D, et al. Treacher Collins syndrome: a clinical and molecular study based on a large series of patients. Genet Med. 2016;18(1):49-56. doi:10.1038/gim.2015.29 PMID: 25790162.
13. Sanchez E, Laplace-Builhé B, Mau-Them FT, Richard E, Goldenberg A, Toler TL, et al. POLR1B and neural crest cell anomalies in Treacher Collins syndrome type 4. Genet Med. 2020;22(3):547-556. doi:10.1038/s41436-019-0669-9 PMID: 31649276. PMCID: PMC7056642.
14. Orphanet. Mandibulofacial dysostosis (ORPHA:155899, group). https://www.orpha.net/en/disease/detail/155899
15. Bernard G, Vanderver A. POLR3-Related Leukodystrophy. 2012 Aug 2 [Updated 2017 May 11]. In: Adam MP, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–. https://www.ncbi.nlm.nih.gov/books/NBK99167/ PMID: 22855961.
16. Orphanet. Hypomyelination-hypogonadotropic hypogonadism-hypodontia syndrome (ORPHA:88637). https://www.orpha.net/en/disease/detail/88637
17. Thiffault I, Wolf NI, Forget D, Guerrero K, Tran LT, Choquet K, et al. Recessive mutations in POLR1C cause a leukodystrophy by impairing biogenesis of RNA polymerase III. Nat Commun. 2015;6:7623. doi:10.1038/ncomms8623 PMID: 26151409. PMCID: PMC4506509.
18. Reid L, Carroll W. Treacher Collins Syndrome in the United States: Examining Incidence and Inpatient Interventions. Cleft Palate Craniofac J. 2021;58(11):1438-1442. doi:10.1177/1055665621998138 PMID: 33784859.
19. Plomp RG, van Lieshout MJS, Joosten KFM, Wolvius EB, van der Schroeff MP, Versnel SL, et al. Treacher Collins Syndrome: A Systematic Review of Evidence-Based Treatment and Recommendations. Plast Reconstr Surg. 2016;137(1):191-204. doi:10.1097/prs.0000000000001896 PMID: 26710023. Supplemental Digital Content (treatment algorithm): http://links.lww.com/PRS/B512
20. Van Roey VL, Irvine WFE, All Expert Panel Members of the Facial Dysostosis Consensus Statement. Optimal Diagnostic and Treatment Practices for Facial Dysostosis Syndromes: A Clinical Consensus Statement Among European Experts. J Craniofac Surg. 2024;35(5):1315-1324. doi:10.1097/scs.0000000000010280 PMID: 38801252. PMCID: PMC11198962. Supplemental Digital Content: http://links.lww.com/SCS/G243 to G247
21. ClinicalTrials.gov. Search "Treacher Collins" (API v2) [Internet]. Bethesda (MD): National Library of Medicine; [accessed 10 Oct 2026]. Available at: https://clinicaltrials.gov/api/v2/studies?query.cond=Treacher+Collins
22. ERN CRANIO. Facial Dysostosis Consensus Statement — Standards of care [Internet]. Rotterdam: European Reference Network for rare and/or complex craniofacial anomalies and ENT disorders; [accessed 10 Oct 2026]. Available at: https://www.ern-cranio.eu/standards-of-care
23. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 232–4 (Micrognathia and Robin sequence).
24. Warner RM, Fagan JM, Das-Gupta R. Treacher Collins syndrome: an undescribed characteristic of the condition and its management with botulinum toxin and surgery. J Craniofac Surg. 2008;19(6):1600-1602. doi:10.1097/scs.0b013e31818ac1d6 PMID: 19098560.
25. The Treacher Collins Syndrome Collaborative Group. Positional cloning of a gene involved in the pathogenesis of Treacher Collins syndrome. Nat Genet. 1996;12(2):130-136. doi:10.1038/ng0296-130 PMID: 8563749.
26. Splendore A, Silva EO, Alonso LG, Richieri-Costa A, Alonso N, Rosa A, et al. High mutation detection rate in TCOF1 among Treacher Collins syndrome patients reveals clustering of mutations and 16 novel pathogenic changes. Hum Mutat. 2000;16(4):315-322. doi:10.1002/1098-1004(200010)16:4<315::AID-HUMU4>3.0.CO;2-H PMID: 11013442.
27. Dauwerse JG, Dixon J, Seland S, Ruivenkamp CA, van Haeringen A, Hoefsloot LH, et al. Mutations in genes encoding subunits of RNA polymerases I and III cause Treacher Collins syndrome. Nat Genet. 2011;43(1):20-22. doi:10.1038/ng.724 PMID: 21131976.
28. Schaefer E, Collet C, Genevieve D, Vincent M, Lohmann DR, Sanchez E, et al. Autosomal recessive POLR1D mutation with decrease of TCOF1 mRNA is responsible for Treacher Collins syndrome. Genet Med. 2014;16(9):720-724. doi:10.1038/gim.2014.12 PMID: 24603435.
29. Ferraro F, Kühn N, Rots D, van der Linde HC, Mohseni B, van Unen L, et al. Long-read DNA and RNA sequencing reveal an intronic retrotransposon insertion in TCOF1 causing Treacher Collins syndrome. HGG Adv. 2026;7(1):100523. doi:10.1016/j.xhgg.2025.100523 PMID: 41017149. PMCID: PMC12816837.
30. Vincent M, Collet C, Verloes A, et al. Large deletions encompassing the TCOF1 and CAMK2A genes are responsible for Treacher Collins syndrome with intellectual disability. Eur J Hum Genet. 2014;22:52-6. doi:10.1038/ejhg.2013.98 PMID: 23695276.
31. GeneGuidance — TCOF1, POLR1D, POLR1C entries (version of 26 Sep 2026).
32. Ferreira CR, Darling A, Vockley J. Disorders of nucleic acid metabolism, tRNA metabolism and ribosomal biogenesis. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 39, section 39.3. doi:10.1007/978-3-662-63123-2_39
33. Marszałek-Kruk BA, Wójcicki P, Dowgierd K, Śmigiel R. Treacher Collins Syndrome: Genetics, Clinical Features and Management. Genes (Basel). 2021;12(9):1392. doi:10.3390/genes12091392 PMID: 34573374. PMCID: PMC8470852.
34. Gonzales B, Henning D, So RB, Dixon J, Dixon MJ, Valdez BC. The Treacher Collins syndrome (TCOF1) gene product is involved in pre-rRNA methylation. Hum Mol Genet. 2005;14:2035-43. doi:10.1093/hmg/ddi208 PMID: 15930015.
35. Dixon J, Jones NC, Sandell LL, Jayasinghe SM, Crane J, Rey JP, et al. Tcof1/Treacle is required for neural crest cell formation and proliferation deficiencies that cause craniofacial abnormalities. Proc Natl Acad Sci U S A. 2006;103(36):13403-13408. doi:10.1073/pnas.0603730103 PMID: 16938878. PMCID: PMC1557391.
36. Dixon J, Brakebusch C, Fassler R, Dixon MJ. Increased levels of apoptosis in the prefusion neural folds underlie the craniofacial disorder, Treacher Collins syndrome. Hum Mol Genet. 2000;9:1473-80. doi:10.1093/hmg/9.10.1473 PMID: 10888597.
37. Jones NC, Lynn ML, Gaudenz K, Sakai D, Aoto K, Rey JP, et al. Prevention of the neurocristopathy Treacher Collins syndrome through inhibition of p53 function. Nat Med. 2008;14(2):125-133. doi:10.1038/nm1725 PMID: 18246078. PMCID: PMC3093709.
38. Sakai D, Dixon J, Achilleos A, Dixon M, Trainor PA. Prevention of Treacher Collins syndrome craniofacial anomalies in mouse models via maternal antioxidant supplementation. Nat Commun. 2016;7:10328. doi:10.1038/ncomms10328 PMID: 26792133. PMCID: PMC4735750.
39. Ni C, Ichino L, Swigut T, Wysocka J. Enhancer buffering protects dosage-sensitive housekeeping genes during vulnerable developmental transitions. bioRxiv [preprint]. 2026. doi:10.64898/2026.07.29.741571
40. Antal G, Zsigmond A, Till Á, Szabó A, Maász A, Bene J, et al. Molecular and Clinical Heterogeneity in Hungarian Patients with Treacher Collins Syndrome-Identification of Two Novel Mutations by Next-Generation Sequencing. Int J Mol Sci. 2024;25(21):11400. doi:10.3390/ijms252111400 PMID: 39518953. PMCID: PMC11546311.
41. Camarillo-Benitez S, Mendez-Catala CF, Chima-Galan MDC, Rivera-Yañez CR, Negrete-Torres N, Arrieta TA, et al. Genetic and Molecular Characterization of Treacher Collins Syndrome in Three Mexican Families. Int J Mol Sci. 2026;27(4):1891. doi:10.3390/ijms27041891 PMID: 41752027. PMCID: PMC12941230.
42. Walker-Kopp N, Jackobel AJ, Pannafino GN, Morocho PA, Xu X, Knutson BA. Treacher Collins syndrome mutations in Saccharomyces cerevisiae destabilize RNA polymerase I and III complex integrity. Hum Mol Genet. 2017;26(21):4290-4300. doi:10.1093/hmg/ddx317 PMID: 28973381.
43. Kawakami K, Kato H. Analysis of Treacher Collins syndrome 4-associated mutations in Schizosaccharomyces pombe. FEBS Open Bio. 2026;16(3):561-569. doi:10.1002/2211-5463.70152 PMID: 41542823.
44. Ulhaq ZS, Nurputra DK, Soraya GV, Kurniawati S, Istifiani LA, Pamungkas SA, et al. A systematic review on Treacher Collins syndrome: Correlation between molecular genetic findings and clinical severity. Clin Genet. 2023;103(2):146-155. doi:10.1111/cge.14243 PMID: 36203321.
45. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 142–4 (Ear anomalies).
46. Teber OA, Gillessen-Kaesbach G, Fischer S, et al. Genotyping in 46 patients with tentative diagnosis of Treacher Collins syndrome revealed unexpected phenotypic variation. Eur J Hum Genet. 2004;12:879-90. doi:10.1038/sj.ejhg.5201260 PMID: 15340364.
47. Zhang L, Wang F, Zhu Y, Zhang H, Liu Y. Case Report: A novel heterozygous variant of the TCOF1 gene in Treacher Collins syndrome. Front Pediatr. 2025;13:1615309. doi:10.3389/fped.2025.1615309 PMID: 41036210. PMCID: PMC12479431.
48. Enomoto Y, Tsurusaki Y, Tominaga M, Kobayashi S, Inoue M, Fujita K, et al. A Recurrent Variant in POLR1B, c.3007C>T; p.Arg1003Cys, Associated with Atresia of the External Canal and Microtia in Treacher Collins Syndrome Type 4. Mol Syndromol. 2021;12(2):127-132. doi:10.1159/000513224 PMID: 34012383. PMCID: PMC8114036.
49. Ghesh L, Vincent M, Delemazure AS, Boyer J, Corre P, Perez F, et al. Autosomal recessive Treacher Collins syndrome due to POLR1C mutations: Report of a new family and review of the literature. Am J Med Genet A. 2019;179(7):1390-1394. doi:10.1002/ajmg.a.61147 PMID: 30957429.
50. Bowman M, Oldridge M, Archer C, et al. Gross deletions in TCOF1 are a cause of Treacher-Collins-Franceschetti syndrome. Eur J Hum Genet. 2012;20:769-77. doi:10.1038/ejhg.2012.2 PMID: 22317976.
51. Dixon J, Ellis I, Bottani A, Temple K, Dixon MJ. Identification of mutations in TCOF1: use of molecular analysis in the pre- and postnatal diagnosis of Treacher Collins syndrome. Am J Med Genet A. 2004;127A:244-8. doi:10.1002/ajmg.a.30010 PMID: 15150774.
52. Fan F, Chen Y, Zhang T, Ma J. Novel TCOF1 Frameshift Variant and Phenotypic Heterogeneity in a Chinese Family With Treacher Collins Syndrome. Mol Genet Genomic Med. 2026;14(9):e70297. doi:10.1002/mgg3.70297 PMID: 42734254. PMCID: PMC13573626.
53. Splendore A, Jabs EW, Félix TM, Passos-Bueno MR. Parental origin of mutations in sporadic cases of Treacher Collins syndrome. Eur J Hum Genet. 2003;11(9):718-722. doi:10.1038/sj.ejhg.5201029 PMID: 12939661.
54. Fitriasari S, Fiorino R, Truong TH, McKinney MC, Dixon J, Dixon MJ, et al. Gene-environment interactions modulate the phenotype severity in mouse models of congenital craniofacial syndromes. J Clin Invest. 2025;135(19):e181705. doi:10.1172/jci181705 PMID: 40699891. PMCID: PMC12483615.
55. Sands LB, Goo CE, White LK, Lubash BT, McKinney MC, Guo F, et al. Loss of polr1c disrupts myelination in a zebrafish model of POLR1C -associated disease. bioRxiv [preprint]. 2026. doi:10.64898/2026.09.24.752997
56. Beaumont CA, Dunaway DJ, Padwa BL, Forrest C, Koudstaal MJ, Caron CJJM. Extracraniofacial anomalies in Treacher Collins syndrome: A multicentre study of 248 patients. Int J Oral Maxillofac Surg. 2021;50(11):1471-1476. doi:10.1016/j.ijom.2021.03.001 PMID: 33752939.
57. Rooijers W, Schreuder MJ, Loudon SE, Wan MJ, Dunaway DJ, Padwa BL, et al. Ocular and adnexal anomalies in Treacher Collins syndrome: a retrospective multicenter study. J AAPOS. 2022;26(1):10.e1-10.e6. doi:10.1016/j.jaapos.2021.10.005 PMID: 35032653.
58. Hertle RW, Ziylan S, Katowitz JA. Ophthalmic features and visual prognosis in the Treacher-Collins syndrome. Br J Ophthalmol. 1993;77(10):642-645. doi:10.1136/bjo.77.10.642 PMID: 8218033. PMCID: PMC504607.
59. Ng JJ, Zhong A, Massenburg BB, Binenbaum G, Wu M, Romeo DJ, et al. Periorbital Outcomes and Vision Risk Stratification in Treacher Collins Syndrome. Plast Reconstr Surg. 2025;156(2):246e-256e. doi:10.1097/prs.0000000000011563 PMID: 38857421.
60. Ansari S, Basappa ER, Markee PC, Dixon C, Bhabad SH. Spectrum of Otological Manifestations in Treacher Collins Syndrome: A Case Series of 9 Patients. J Comput Assist Tomogr. 2025;49(4):662-668. doi:10.1097/rct.0000000000001715 PMID: 39876535.
61. Guo Q, Zhu J, Wang D, Zhao S. Hearing rehabilitation in patients with Treacher Collins syndrome. Acta Otolaryngol. 2026;146(5):645-653. doi:10.1080/00016489.2026.2623955 PMID: 41949491.
62. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 122–5 (Deafness in early childhood).
63. Wu M, Barnett SL, Massenburg BB, Friedman LM, Ng JJ, Romeo DJ, et al. Clarifying the Relationship Between Orbito-Zygomatic and Mandibular Dysmorphology in Treacher Collins Syndrome. J Craniofac Surg. 2025;36(3):887-891. doi:10.1097/scs.0000000000010881 PMID: 40558009.
64. Plomp RG, Bredero-Boelhouwer HH, Joosten KF, Wolvius EB, Hoeve HL, Poublon RM, et al. Obstructive sleep apnoea in Treacher Collins syndrome: prevalence, severity and cause. Int J Oral Maxillofac Surg. 2012;41(6):696-701. doi:10.1016/j.ijom.2012.01.018 PMID: 22521672.
65. Akre H, Øverland B, Åsten P, Skogedal N, Heimdal K. Obstructive sleep apnea in Treacher Collins syndrome. Eur Arch Otorhinolaryngol. 2012;269(1):331-337. doi:10.1007/s00405-011-1649-0 PMID: 21626120. PMCID: PMC3249173.
66. Paradowska-Stolarz A, Sluzalec-Wieckiewicz K, Mikulewicz M, Maslowiec C, Kokot K, Miralles-Jorda L, et al. Obstructive Sleep Apnoea in Patients with Treacher Collins Syndrome-A Narrative Review. J Clin Med. 2025;14(13):4741. doi:10.3390/jcm14134741 PMID: 40649114. PMCID: PMC12251209.
67. Østertun Geirdal A, Øverland B, Heimdal K, Storhaug K, Asten P, Akre H. Association between obstructive sleep apnea and health-related quality of life in individuals affected with Treacher Collins syndrome. Eur Arch Otorhinolaryngol. 2013;270(11):2879-2884. doi:10.1007/s00405-013-2409-0 PMID: 23455582.
68. Østerhus IN, Skogedal N, Akre H, Johnsen UL, Nordgarden H, Åsten P. Salivary gland pathology as a new finding in Treacher Collins syndrome. Am J Med Genet A. 2012;158A(6):1320-1325. doi:10.1002/ajmg.a.35331 PMID: 22585367.
69. Asten P, Skogedal N, Nordgarden H, Axelsson S, Akre H, Sjögreen L. Orofacial functions and oral health associated with Treacher Collins syndrome. Acta Odontol Scand. 2013;71(3-4):616-625. doi:10.3109/00016357.2012.700065 PMID: 22783882.
70. da Silva Dalben G, Costa B, Gomide MR. Prevalence of dental anomalies, ectopic eruption and associated oral malformations in subjects with Treacher Collins syndrome. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2006;101(5):588-592. doi:10.1016/j.tripleo.2005.07.016 PMID: 16632269.
71. Medeiros LH, Bueno PM, Fidelis da Silva LV, Fukushiro AP, Trindade-Suedam IK. Evaluation of masticatory function in the pediatric and adult populations with Treacher Collins syndrome. J Oral Biol Craniofac Res. 2025;15(6):1460-1466. doi:10.1016/j.jobcr.2025.08.033 PMID: 40989145. PMCID: PMC12452860.
72. Giabicani E, Lemale J, Dainese L, et al. Chronic intestinal pseudo-obstruction in a child with Treacher Collins syndrome. Arch Pediatr. 2017;24:1000-4. doi:10.1016/j.arcped.2017.07.004 PMID: 28927774.
73. Lines M, Hartley T, MacDonald SK, Boycott KM. Mandibulofacial Dysostosis with Microcephaly. 2014 Jul 3 [Updated 2023 Apr 6]. In: Adam MP, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–. https://www.ncbi.nlm.nih.gov/books/NBK214367/
74. Wenger TL, Wild KT, Zaniletti I, Zackai EH, Lioy J, Resnick CM, et al. Management and Outcomes of Neonates with Treacher Collins and Nager Syndromes. J Pediatr. 2025;283:114614. doi:10.1016/j.jpeds.2025.114614 PMID: 40280475.
75. Wang Y, Wang Y, Yao M, Chen L, Wu S, Liu Y. Prenatal diagnosis of Treacher Collins syndrome: a case report and literature review. Int J Gynaecol Obstet. 2023;163:778-81. doi:10.1002/ijgo.14881 PMID: 37231986.
76. Zhu H, Du M, Zhu S, Huang Y, Zeng L, Kuai Y, et al. A novel pathogenic variant in POLR1D (c.220dup, p.His74ProfsTer8) causes Treacher Collins syndrome type 2 in a Chinese patient: a case report. BMC Pediatr. 2025;26(1):68. doi:10.1186/s12887-025-06469-9 PMID: 41455926. PMCID: PMC12853992.
77. Sng JHT, Hue JJJ, Rubio-Palau J. Evaluation of a Facial Dysmorphology Analysis Algorithm (Face2Gene) in Identifying Treacher Collins Syndrome Amongst Diverse Population. Orthod Craniofac Res. 2026;29(4):680-687. doi:10.1111/ocr.70113 PMID: 41817050. PMCID: PMC13485199.
78. Pereira DC, Bussamra LC, Araujo Júnior E, Drummond CL, Nardozza LM, Moron AF, et al. Prenatal diagnosis of treacher-collins syndrome using three-dimensional ultrasonography and differential diagnosis with other acrofacial dysostosis syndromes. Case Rep Obstet Gynecol. 2013;2013:203976. doi:10.1155/2013/203976 PMID: 23653874. PMCID: PMC3638565.
79. Wang X, Jia F, Zhang B, Liu S, Wang H, Chen J, et al. Fetal Phenotype and Diagnosis of Treacher Collins Syndrome Due to a Novel POLR1D Variant. Prenat Diagn. 2026;46(1):138-144. doi:10.1002/pd.70035 PMID: 41351247.
80. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 364–5 (Mandibulofacial dysostosis with microcephaly).
81. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 370–1 (Miller syndrome).
82. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 372–3 (Nager syndrome).
83. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 886–8 (Nager syndrome).
84. Orphanet. Mandibulofacial dysostosis-microcephaly syndrome (ORPHA:79113). https://www.orpha.net/en/disease/detail/79113
85. Orphanet. Nager syndrome (ORPHA:245). https://www.orpha.net/en/disease/detail/245
86. Orphanet. Postaxial acrofacial dysostosis (ORPHA:246). https://www.orpha.net/en/disease/detail/246
87. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 296–9 (Radial ray defects and thumb hypoplasia).
88. Duley JA, Henman MG, Carpenter KH, Bamshad MJ, Marshall GA, Ooi CY, et al. Elevated plasma dihydroorotate in Miller syndrome: Biochemical, diagnostic and clinical implications, and treatment with uridine. Mol Genet Metab. 2016;119(1-2):83-90. doi:10.1016/j.ymgme.2016.06.008 PMID: 27370710.
89. Weaver KN, Watt KE, Hufnagel RB, et al. Acrofacial dysostosis, Cincinnati type, a mandibulofacial dysostosis syndrome with limb anomalies, is caused by POLR1A dysfunction. Am J Hum Genet. 2015;96:765-74. doi:10.1016/j.ajhg.2015.03.011 PMID: 25913037.
90. Silvey S, Lovell S, Butler MG. RNA Polymerase I Dysfunction Underlying Craniofacial Syndromes: Integrated Genetic Analysis Reveals Parallels to 22q11.2 Deletion Syndrome. Genes (Basel). 2025;16(9):1063. doi:10.3390/genes16091063 PMID: 41010008. PMCID: PMC12470080.
91. Ulhaq ZS, Tse WKF. Comparative Transcriptomics Reveals Shared Downstream Pathways in Craniofacial Pathology. Cleft Palate Craniofac J. 2026. Online ahead of print. doi:10.1177/10556656261485119 PMID: 42709501.
92. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 146–9 (Facial asymmetry).
93. Singh DJ, Bartlett SP. Congenital mandibular hypoplasia: analysis and classification. J Craniofac Surg. 2005;16:291-300. doi:10.1097/00001665-200503000-00017 PMID: 15750428.
94. Ribeiro EM, coord. Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. p. 197–9 (10.3 Sequência de Pierre Robin, Quadro 10.3.3).
95. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 110–1 (Coloboma).
96. American Cleft Palate-Craniofacial Association. Parameters For Evaluation and Treatment of Patients With Cleft Lip/Palate or Other Craniofacial Differences. Cleft Palate Craniofac J. 2018;55(1):137-156. doi:10.1177/1055665617739564 PMID: 34162066.
97. Thompson JT, Anderson PJ, David DJ. Treacher Collins syndrome: protocol management from birth to maturity. J Craniofac Surg. 2009;20:2028-35. doi:10.1097/scs.0b013e3181be8788 PMID: 19881372.
98. Trainor PA, Andrews BT. Facial dysostoses: etiology, pathogenesis and management. Am J Med Genet C Semin Med Genet. 2013;163C:283-94. doi:10.1002/ajmg.c.31375 PMID: 24123981.
99. Dounas GD, Farmer LDM, Kwon HJ, Moore MH. Mandibular Lengthening by Distraction for Airway Obstruction in Treacher-Collins Syndrome: The 30-Year Results. J Craniofac Surg. 2025. Online ahead of print. doi:10.1097/scs.0000000000011255 PMID: 40172972.
100. Lei B, Huang Q, Sun T, Zhong Y, Chen Z, Yang B. Comprehensive serial treatment of Treacher Collins syndrome. J Craniofac Surg. 2023;34:e102-e104. doi:10.1097/scs.0000000000008772 PMID: 35883237.
101. Kobus K, Wojcicki P. Surgical treatment of Treacher Collins syndrome. Ann Plast Surg. 2006;56:549-54. doi:10.1097/01.sap.0000196580.56569.d2 PMID: 16641634.
102. Marinac I, Trotić R, Košec A. Systematic Review of Current Audiological Treatment Options for Patients with Treacher Collins Syndrome (TCS) and Surgical and Audiological Experiences of an Otorhinolaryngologist with TCS. J Pers Med. 2024;14(1):81. doi:10.3390/jpm14010081 PMID: 38248782. PMCID: PMC10817470.
103. Walker BM, Zimmerman CN, Caruth KM, Knutson BA, Palumbo RJ. Maternal antioxidant treatment partially rescues developmental defects in a Drosophila Polr1D mutant model. MicroPubl Biol. 2025;2025. doi:10.17912/micropub.biology.001634 PMID: 40583963. PMCID: PMC12203242.
104. U.S. Food and Drug Administration. Premarket Notification 510(k) K984162 (paediatric BAHA), K183373 (Bonebridge), K190589 and K260902 (Osia), K261142 (Ponto) e De Novo DEN170009 (Bonebridge): decisions, summaries and indications for use [Internet]. Silver Spring (MD): FDA; [accessed 10 Oct 2026]. Available at: https://www.accessdata.fda.gov/cdrh_docs/ and https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/denovo.cfm?id=DEN170009
105. European Commission. EUDAMED — European Database on Medical Devices: public device search (trade names Osia, Ponto, BI300, Baha, Bonebridge) [Internet]. Brussels: European Commission; [accessed 10 Oct 2026]. Available at: https://ec.europa.eu/tools/eudamed
106. van Roey VL, Ombashi S, Mathijssen IMJ, Munkhammar ÅA, Åsten PM, Bouzariouh A, et al. The Development of a European Registry for Facial Dysostosis Syndromes: A Delphi-Guided Approach. J Craniofac Surg. 2025;36(8):2712-2716. doi:10.1097/scs.0000000000011695 PMID: 40694792. PMCID: PMC12537031.
107. Zhang X, Liu X, Duan L, Liu X, Chen Y, Feng X, et al. Pregnancy with treacher collins syndrome in mechanical heart valve: a case report and discussion treacher collins syndrome and MHV. BMC Pregnancy Childbirth. 2026;26(1):630. doi:10.1186/s12884-026-09042-z PMID: 42032534. PMCID: PMC13244981.
108. Trainor PA, Dixon J, Dixon MJ. Treacher Collins syndrome: etiology, pathogenesis and prevention. Eur J Hum Genet. 2009;17:275-83. doi:10.1038/ejhg.2008.221 PMID: 19107148.
109. Marres HA. Hearing loss in the Treacher-Collins syndrome. Adv Otorhinolaryngol. 2002;61:209-15. doi:10.1159/000066811 PMID: 12408086.
110. Shoo BA, McPherson E, Jabs EW. Mosaicism of a TCOF1 mutation in an individual clinically unaffected with Treacher Collins syndrome. Am J Med Genet A. 2004;126A:84-8. doi:10.1002/ajmg.a.20488 PMID: 15039977.
111. Montgomery K, Harcourt D. Reproductive decision-making in the context of inheritable conditions that affect appearance: Lived experiences and perspectives. J Community Genet. 2026;17(4):92. doi:10.1007/s12687-026-00926-3 PMID: 42533209. PMCID: PMC13424007.
112. Laposa — Landelijke Patiënten- en Oudervereniging voor Schedel- en/of Aanzichtsaandoeningen [Internet]. Netherlands; [accessed 10 Oct 2026]. Available at: https://www.laposa.nl/
113. Changing Faces. Treacher Collins syndrome — condition-specific information [Internet]. London: Changing Faces; [accessed 10 Oct 2026]. Available at: https://www.changingfaces.org.uk/advice-guidance/condition-specific-information/treacher-collins-syndrome/
114. Treacher Collins Family Support Group (registered charity 1006300) [Internet]. London: Charity Commission for England and Wales; [accessed 10 Oct 2026]. Available at: https://register-of-charities.charitycommission.gov.uk/en/charity-search/-/charity-details/1006300
115. Children's Craniofacial Association [Internet]. Dallas (TX): CCA; [accessed 10 Oct 2026]. Available at: https://ccakids.org/