Tourette Syndrome (TS) is a complex neurodevelopmental disorder characterized by persistent motor and vocal tics with an onset in childhood. It affects approximately 1% of the general population, with a male-to-female ratio of 3-4:1 (1–3).

The etiology of TS is multifactorial, involving a dynamic interplay between genetic, epigenetic, and environmental factors (4, 5). Twin and family studies support a strong heritable component and suggest that TS ranks among the neurodevelopmental disorders with the highest non-Mendelian transmission (6, 7).

Advances in next-generation sequencing, particularly whole-exome sequencing (WES), have substantially enhanced our understanding of the genetic architecture of TS, enabling the identification of both de novo and inherited rare coding variants that disrupt key neurodevelopmental pathways (8). These exomic variants include loss-of-function (LoF) and deleterious missense mutations in genes that are highly constrained and intolerant to variation (9).

Several WES studies in simplex trios have reported damaging de novo variants in approximately 10% of TS cases, with high-confidence risk genes such as CELSR3, WWC1, NIPBL, FN1, and ASH1L emerging as relevant contributors (8, 10). Beyond these core genes, recent research has highlighted the potential contribution of rare variants in candidate genes such as PNKD and BRPF1, which play key roles in presynaptic regulation and chromatin-mediated transcriptional control, respectively (11, 12). Variants in these genes may disrupt synaptic vesicle dynamics or epigenetic modulation, both of which are essential for maintaining the functional integrity of cortico-striatal-thalamo-cortical circuits implicated in tic generation. Moreover, WES meta-analyses have reported an enrichment of X-linked and autosomal rare variants, helping to explain the sex bias observed in TS and its frequent comorbidity with ASD and ADHD (8, 13).

Among these findings, a growing number of variants of uncertain significance (VUS), mutations for which the pathogenic potential remains unconfirmed, are being recognized as crucial to the expanding landscape of TS genetics (14). Despite their classification uncertainty, many VUS occur in genes with plausible neurobiological roles, such as those involved in ion channel activity (e.g., CACNA1D, SLC6A1) (15), GABAergic inhibition (16), chromatin remodeling, and neuronal migration (e.g., RELN, ZMYM2, KMT2C) (17–19). Importantly, some of these VUS may act as susceptibility loci or modifiers within a broader oligogenic or polygenic context, interacting with other rare or common variants to influence phenotypic expression and disease severity. Therefore, careful interpretation of VUS, especially those of de novo origin or affecting functionally constrained regions, is essential for refining the genotype–phenotype correlations in TS.

Building on this background, the present study aimed to clarify whether distinct genetic signatures can help stratify TS patients into clinically meaningful subtypes, with potential implications for early diagnosis and tailored interventions. In particular, we investigated whether potentially pathogenic exomic variants, including de novo variants and VUS, were associated with recurrent phenotypic features in a clinically well-characterized pediatric TS cohort.

By systematically analyzing the clinical and molecular profiles of affected individuals, we sought to identify genetic landmarks that may underlie distinct clinical subtypes of TS, contributing to the broader understanding of its heterogeneity and informing future precision medicine approaches.

By integrating genetic findings with detailed neuropsychiatric assessments, this study aimed to provide novel insights into the contribution of rare exomic variants, including both de novo and inherited mutations, to the clinical heterogeneity of Tourette syndrome.

Our cohort of 80 children and adolescents with TS was characterized by a predominance of males (male-to-female ratio 7:1) and a mean age of 12.8 years, in line with epidemiological data showing the higher prevalence and earlier onset of tics in boys (2, 3).

From a neurobiological and genetic perspective, sex differences in TS and related neurodevelopmental disorders have been interpreted within the framework of sex-specific liability thresholds and the female protective effect. According to this model, females may require a higher cumulative genetic burden to manifest clinically overt disease, a mechanism supported across multiple neurodevelopmental conditions and increasingly suggested by exome-based studies in tic-related phenotypes (8).

Across the sample, tics ranged from mild to severe, and psychiatric comorbidities such as obsessive–compulsive disorder (OCD), attention-deficit/hyperactivity disorder (ADHD), anxiety, and oppositional defiant disorder (ODD) were frequent, reflecting the well-established clinical spectrum of TS (38). Dysmorphic traits, intellectual disability, and epilepsy were also observed in a relevant proportion of patients, especially those carrying potentially causative variants (PC-Vs). MRI anomalies were present in about a quarter of the cohort, and EEG abnormalities in around 10–20%, findings consistent with the variable but nonspecific neurological alterations described in the literature (39).

Genetic subgrouping revealed clear clinical distinctions. Children with PC-Vs displayed the highest tic severity scores on the Yale Global Tic Severity Scale (YGTSS), along with significantly lower IQ values compared to those with VUS or non-causative variants. In contrast, patients without candidate variants more often reported a positive family history of tics, supporting the hypothesis that inherited polygenic burden drives familial clustering, while de novo high-effect variants underlie more severe simplex and syndromic presentations. These demographic and clinical patterns reinforce the dual model of TS genetic architecture, in which rare variants act as strong risk factors for severe phenotypes, and common variants contribute to broader familial risk (8, 40, 41). Importantly, the classification of variants into potentially causative, uncertain, or non-causative categories was not based solely on gene–disease associations but relied on an integrative framework combining variant-level features (including variant type, population frequency, and in silico pathogenicity predictions) with gene-level evidence. This approach was adopted to reduce misclassification and to account for the heterogeneous nature of rare genetic variation in neurodevelopmental disorders.

Building on these observations, the detailed analysis of individual PC-V cases provided further insight into how specific genes contribute to tic severity, cognitive outcome, and associated comorbidities.

Patient 1, carrying a de novo truncating variant in KMT2C, presented with borderline intellectual functioning, dysmorphic features, ADHD, and Tourette syndrome. Pathogenic KMT2C variants have been associated with developmental delay and dysmorphism (42, 43), but tics have not been systematically described, suggesting that our case expands the known phenotype. Patient 2, harboring a truncating variant in PNKD, showed severe tics with OCD and anxiety but preserved cognition, a pattern strikingly similar to the familial cases reported by Sun et al. (11).

By contrast, Patient 4, with a missense variant in the same gene, displayed only moderate tics without comorbidities, exemplifying the uncertainty in interpreting non-truncating variants in synaptic genes.

Variants in chromatin regulators also emerged as relevant contributors. Patient 3, with a de novo BRPF1 variant, presented with ADHD and tics but preserved cognition and no dysmorphism, diverging from the more syndromic presentations described in the literature (12). Patient 5 carried a SMARCA2 missense variant, typically linked to Nicolaides–Baraitser syndrome with severe intellectual disability and seizures (44), but in our case associated with preserved IQ, a learning disorder, and moderate tics. These cases illustrate how chromatin-remodeling genes, while generally linked to syndromic neurodevelopmental disorders, may predispose to tic phenotypes in milder allelic presentations.

Some patients presented complex phenotypes involving multiple or pleiotropic gene effects. Patient 6 carried dual variants in CACNA1D and RERE and exhibited developmental delay, epilepsy, and tics. Both genes are individually implicated in neurodevelopmental disorders (45, 46), and their co-occurrence suggests additive or oligogenic effects, as previously hypothesized in TS cohorts with multiple rare variants (40).

Similarly, Patient 7 carried a de novo variant in MAPK8IP3, a gene associated with intellectual disability and axonal transport dysfunction (47). Although tics have not been reported in MAPK8IP3-related disorders, their presence in our case suggests that axonal transport defects may also predispose to movement disorders, extending the clinical spectrum of this condition.

Other findings highlighted interpretative challenges. Patient 8 harbored a heterozygous missense variant in SLC20A2, classically linked to adult-onset primary familial brain calcification (48). In this child, the absence of calcifications renders the variant more consistent with a VUS, analogous to CNV findings in pediatric TS where variants in late-onset genes are difficult to classify. In contrast, Patient 9 carried a de novo PTEN nonsense mutation and exhibited macrocephaly, dysmorphism, intellectual disability, ADHD, and tics, a constellation highly consistent with PTEN hamartoma tumor syndrome (49). Patient 10, with a de novo ANKRD17 variant, showed dysmorphism, borderline cognition, and tics. While ANKRD17 has been linked to intellectual disability and behavioral disturbances (50), tics have not been reported, suggesting an expansion of the phenotypic spectrum.

Finally, Patient 11 carried a de novo missense variant in SLC6A1 and presented with epilepsy, ADHD, OCD, anxiety, and tics. Pathogenic SLC6A1 variants are well-recognized causes of epilepsy with developmental delay and psychiatric comorbidities (51). The strong phenotypic overlap with published cases supports impaired GABAergic inhibition as a convergent mechanism in Tourette syndrome (40).

Taken together, these cases highlight recurring themes: truncating and de novo variants in constrained genes are associated with more severe and syndromic phenotypes, while inherited or missense variants remain more challenging to interpret but may still act as modifiers. The overlap with genes implicated in broader neurodevelopmental syndromes underscores the pleiotropy and shared vulnerability across psychiatric and neurodevelopmental disorders. Psychiatric comorbidities such as OCD, ADHD, and anxiety were frequent in all groups but did not significantly differ by variant type, suggesting they are more strongly influenced by polygenic and environmental factors (38). Likewise, MRI and EEG anomalies did not differentiate between genetic subgroups, consistent with clinical guidelines that do not recommend routine imaging in typical TS (39).

Our findings underscore the complementary contributions of rare high-effect variants and inherited polygenic burden in the pathogenesis of Tourette syndrome. Patients with potentially pathogenic variants tend to exhibit higher tic severity, lower cognitive scores, and syndromic features, while familial cases without such variants reflect polygenic inheritance. Comprehensive genomic testing is particularly warranted in patients with atypical or syndromic features, while future studies integrating rare and common variant analyses with longitudinal phenotyping and functional validation will be critical to refine variant interpretation, improve classification frameworks, and develop mechanism-driven interventions.

This study explored the genetic basis and clinical heterogeneity of Tourette Syndrome in a pediatric cohort by integrating clinical exome sequencing with detailed phenotypic evaluation. The findings confirm the dual contribution of genetic mechanisms: on one side, rare, high-impact variants were associated with more severe clinical outcomes, including higher tic severity, lower cognitive scores, and syndromic features; on the other, the absence of clearly pathogenic variants correlated more strongly with familial aggregation, suggesting a polygenic or inherited background.

From a molecular perspective, the study highlighted the involvement of diverse neurodevelopmental pathways, including chromatin remodeling, synaptic transmission and inhibitory signaling, ion channel function and pleiotropic regulators. These results reinforce the overlap between TS and other neurodevelopmental disorders and support the notion of a shared genetic vulnerability across conditions.

Despite these insights, several limitations should be acknowledged. The relatively small sample size limits the statistical power and generalizability of the findings. Moreover, the classification of variants, particularly variants of uncertain significance, remains challenging and may evolve as new functional and population data become available. The lack of systematic parental testing in all cases also restricted the interpretation of inheritance patterns in some families.

From a clinical standpoint, the study underscores the diagnostic utility of exome sequencing in complex or atypical cases of TS. Identifying rare, potentially causative variants may help guide patient management, genetic counseling, and risk stratification, especially in children presenting with intellectual disability, dysmorphic traits, epilepsy, or structural brain anomalies. At the same time, recognition of polygenic inheritance in familial cases emphasizes the importance of considering both rare and common variants in clinical practice.