Patients visiting the Department of Neurology at Hebei Medical University’s Third Hospital between January 2014 and January 2025 with an initial diagnosis of movement disorders were recruited, and their clinical and genetic data were collected. Participants were stratified into two groups: hypokinetic movement disorders (parkinsonian phenotypes) and hyperkinetic movement disorders groups.

Inclusion criteria for hypokinetic movement disorders group: early-onset parkinsonian syndromes (age of onset ≤50 years), including idiopathic Parkinson’s disease, or ≥1 cardinal parkinsonian feature (bradykinesia, resting tremor, rigidity, or postural instability); inclusion criteria for hyperkinetic movement disorders group: early-onset hyperkinetic disorders (age of onset ≤50 years), including dystonia, chorea, tic disorders, myoclonus. Exclusion Criteria: (1). Secondary causes of parkinsonism or involuntary movements (e.g., infections, neoplasms, stroke, inflammatory demyelination, metabolic disorders); (2). Individuals with probable or possible multiple system atrophy.

Positive familial history: ≥1 first- or second-degree relative with ataxia or movement disorders; sporadic cases: no affected first- or second-degree relatives. Asymptomatic family members underwent genetic testing when DNA was available. The study protocol was approved by the Ethics Committee of the Hebei Medical University Third Hospital. All participants provided written informed consent. Genomic DNA was extracted from peripheral blood using the QIAAmp DNA Blood Mini Kit (QIAGEN, Germany).

Demographic and clinical characteristics of probands were comprehensively analyzed, including gender, family history, age of onset, disease duration, initial symptoms, clinical manifestations and signs (all patients underwent systematic clinical examinations and evaluations by at least two experienced neurologists), and available genetic test results (including existing familial genetic data). Clinical neurological and cognitive function assessments were performed, encompassing brain magnetic resonance imaging (MRI), electromyography (EMG), somatosensory evoked potentials (SEP), the International Cooperative Ataxia Rating Scale (ICARS), the Unified Parkinson’s Disease Rating Scale (UPDRS), the Yale Global Tic Severity Scale (YGTSS), the Mini-Mental State Examination (MMSE), and the Montreal Cognitive Assessment (MoCA).

All patients initially underwent targeted next-generation sequencing (NGS) panel testing (MyGenotics Co., Ltd., Beijing, China), which included genes associated with various movement disorders such as Parkinson’s disease, dystonia, hereditary spastic paraplegia, chorea, and Wilson’s disease. The complete list of genes covered by the targeted NGS panel is s shown in the Supplementary Table S1. The sequence data were mapped using the BWA¹ and SAMTOOLS software^2,3 onto the hg19 human genome as a reference. The variants were identified using the wANNOVAR tool,⁴ and their potential pathogenicity was predicted via the REVEL tool.⁵ Pathogenicity classifications were determined according to the American College of Medical Genetics and Genomics (ACMG) guidelines (27). This targeted NGS panel analysis enabled systematic exclusion of hereditary movement disorders, such as hereditary Parkinson’s disease, hereditary spastic paraplegia, Wilson’s disease, and other movement disorders.

Patients with negative findings on targeted NGS panel testing underwent SCA repeat expansion panel sequencing. This panel included 12 genes: ATXN1, ATXN2, ATXN3, CACNA1A, ATXN7, ATXN8OS/ATXN8, PPP2R2B, TBP, ATN1, FXN, C9orf72, and HTT. The analysis employed fluorescence-labeled PCR followed by capillary electrophoresis (Applied Biosystems^™ 3130xl DNA Analyzers, Thermo Fisher Scientific) to detect pathogenic nucleotide repeat expansions, and molecular weights were determined using GeneMarker software (Promega).

Four probands presenting with parkinsonian phenotypes were identified, including two cases of SCA8, one case of SCA2, and one case of SCA3. The age at onset ranged from 36 to 50 years. All probands exhibited bradykinesia and limb rigidity as initial symptoms, with two cases manifesting unilateral rigidity and the remaining two demonstrating bilateral lower limb rigidity. Proband 1 (F1: II-4), diagnosed with SCA2, harbored a repeat number of 35. She developed parkinsonian symptoms at age 37, followed by cerebellar ataxia 14 years later. Brain MRI revealed mild cerebellar atrophy. Her mother (F1: I-2) and brother exhibited cerebellar ataxia without parkinsonism: the mother developed gait instability around age 30, became wheelchair-bound in her 50s, demonstrated cerebellar atrophy on Brain MRI, and died at 73. This SCA2 pedigree exhibited phenotypic heterogeneity (Table 1; Figure 1B). To date, the remaining three probands (F2: II-1, F4: II-5, F5: II-4) have not developed cerebellar ataxia during follow-up, including one individual with a disease duration of 10 years. Probands (F4: II-5) exhibited gaze-evoked horizontal nystagmus. Three probands (F1: II-4, F4: II-5, F5: II-4) underwent levodopa therapy, with two demonstrating response to levodopa treatment while one showing no significant clinical response to levodopa treatment (Table 2).

Family 4 (F4) represents an SCA8-associated familial parkinsonism pedigree. Affected members exhibited limb rigidity, bradykinesia, and absence of cerebellar ataxia or symptom fluctuations. Besides the proband, two additional affected individuals displayed resting hand tremor. Intriguingly, while proband 4 (F4: II-5) and his elder brother (F4: II-1) harbored similar (CTA/CTG)n repeat expansions (91 vs. 98 repeats), their clinical manifestations diverged. The proband’s brother and father manifested later-onset (post-50 years), levodopa-responsive parkinsonism. The father (F4: I-1) remained ambulatory until his death at age 70. The brother developed involuntary movements in the right upper limb during gait, which resolved with levodopa therapy. In contrast to other family members, the proband exhibited earlier disease onset at age 36, with significantly more severe bradykinesia and limb rigidity. Beyond parkinsonian features, he developed spastic paraparesis of the lower limbs, there are no autonomic dysfunction manifestations (e.g., lower urinary tract dysfunction, orthostatic hypotension). Neurological examination revealed masked facies, marked axial and appendicular bradykinesia, and a combined festinating and spastic gait characterized by forward-flexed posture. Hypertonia with ankle clonus and bilateral positive Chaddock signs were observed in the lower extremities. Although cerebellar ataxia was absent, gaze-evoked horizontal nystagmus was noted. By 10 years post-onset, he experienced profound gait impairment with frequent falls. Brain MRI demonstrated mild cortical atrophy (Figure 1B). Levodopa therapy failed to ameliorate motor symptoms.

Proband 5 (F5: II-4) was diagnosed with a sporadic case of SCA8, the number of (CTA/CTG)n repeat expansion was 55. The patient presented with parkinsonism at age 50, characterized by unilateral bradykinesia and rigidity as initial symptoms, in the absence of tremor or pyramidal signs. Levodopa therapy demonstrated symptomatic improvement.

Proband 3 (F3: III-3) presented at age 35 with multifocal motor and phonic tics, including intermittent facial grimacing, eye blinking, head jerking, arm elevation, chest thrusting, and abdominal contractions, accompanied by repetitive throat-clearing sounds. Each tic lasted several seconds, occurring multiple times daily with exacerbation during emotional arousal and complete resolution during sleep, presenting with obsessions and compulsions, diagnosed with Tourette syndrome with comorbid obsessive-compulsive disorder (OCD). Proband 3 (F3: III-3) developed the aforementioned symptoms at age 10, with progressive worsening over time. Although cerebellar ataxia was absent, the patient had a family history of autosomal dominant cerebellar ataxia: her maternal uncle (F3: II-5) was genetically confirmed with SCA3, and her mother (F3: II-4) had previously presented to our institution with limb ataxia and was diagnosed with SCA3. Given this SCA3-positive familial background, genetic testing was performed, revealing a pathogenic ATXN3 CAG repeat expansion of 62 units. Notably, the proband’s maternal aunt (F3: II-8) presented to our clinic at age 45 with an abnormal crouch-based gait requiring squatting for ambulation, in the absence of cerebellar ataxia or other cerebellar signs, she was diagnosed with somatic symptom disorder (SSD) and anxiety. However, her genetic status remains uncertain as she declined ATXN3 testing (Figure 1; Table 1).

Proband 6 (F6: II-1) developed lower limb spasticity at age 27, predominantly affecting the right leg, manifesting as mild gait disturbance and difficulty flexing the lower limbs. Symptoms remained stable until age 36, when gait impairment progressed with new-onset involuntary movements of both feet. Neurological examination demonstrated a spastic gait in the right lower extremity during ambulation, involuntary right foot inversion during ambulation, bilateral Achilles tendon contractures and pes cavus, and preserved muscle strength (grade 5/5). Hypertonia was observed in both lower limbs (right > left), accompanied by dystonic foot posturing. Bilateral Babinski signs and left Hoffmann sign were present.

The proband’s father exhibited no neurological abnormalities, while his mother died of rectal cancer at age 39 without prior neurological symptoms. Serum ceruloplasmin and homocysteine levels were within normal ranges. Brain MRI showed no structural abnormalities, and lower limb electromyography detected no neurogenic or myopathic changes. Genetic testing identified a TBP CAG/CAA repeat expansion of 41 repeats. Targeted NGS excluded other inherited disorders potentially causative for the clinical phenotype, and no pathogenic heterozygous variants were detected in STUB1. Oral baclofen therapy partially alleviated spasticity but had no effect on involuntary movements.

The global prevalence of SCA8 is relatively low (18, 28–31), and its clinical characteristics have primarily been summarized from small-scale studies, likely attributable to its low disease frequency. Phenotypic heterogeneity among SCA8 patients has been documented across different regions. SCA8 predominantly manifests as slowly progressive cerebellar ataxia (28, 32), but it may also co-occur with or present as other non-ataxic disorders, including paroxysmal kinesigenic dyskinesia (PKD) (18), Parkinson’s disease (11, 20), progressive supranuclear palsy (PSP) (33), Alzheimer’s disease (34), and amyotrophic lateral sclerosis (ALS) (35). In mainland China, cases of SCA8 presenting as Parkinson’s disease are rare, with familial Parkinson’s disease manifestations being exceptionally uncommon. To date, only one such case has been reported by Wang et al. in 2025, describing a patient with SCA8 who exhibited parkinsonian features and responded favorably to levodopa therapy (36).

We identified a novel clinical constellation and significant intrafamilial heterogeneity in SCA8-associated parkinsonism: PD-like phenotype with spastic paraplegia, and levodopa-responsive parkinsonism with dystonia, with detailed discussions as follows:

Proband 4 (F4: II-5) exhibits a PD-like phenotype with spastic paraplegia. He harbored CTA/CTG repeat numbers comparable to his elder brother (91 vs. 98 repeats), yet exhibited distinct phenotypic features. First, the proband 4 exhibited earlier disease onset and more severe parkinsonism. Additionally, severe spastic paraparesis and gaze-evoked nystagmus were documented. Crucially, whereas the proband’s brother (F4: II-1) and father (F4: I-1) demonstrated significant levodopa responsiveness, the proband proved refractory to levodopa therapy. These findings indicate that in addition to repeat length, there may be other factors influencing the phenotype and age of onset of SCA8. Genetic anticipation was evident in proband 4, with disease onset occurring >10 years earlier than his father. Regrettably, ATXN8OS CTA/CTG repeat expansion data were unavailable for the deceased father, precluding confirmation of whether anticipation correlated with repeat length.

The elder brother of proband 4 (F4: II-1) presented with levodopa-responsive parkinsonism and dystonia—a phenotypic combination previously unreported in SCA8;however, Parkinson’s disease with dystonia is relatively common in hereditary Parkinson’s disease. One study showed that foot dystonia is a common manifestation (40%) and occasionally the initial symptom in PD patients harboring Parkin mutations (24). Another study reported cervical dystonia in 9.3% of PD patients, with partial improvement after levodopa therapy (37). A Korean case described levodopa-responsive parkinsonism and mild cerebellar ataxia in SCA8; the proband’s sibling also developed lower-limb dystonia superimposed on parkinsonism and ataxia, while dystonia showed no improvement with levodopa (23). Unlike the case in Korea, our case parallels hereditary PD with dystonia, as levodopa ameliorated both parkinsonism and dystonia. This observation suggests that dystonia in SCA8 presenting as levodopa-responsive parkinsonism may not be coincidental, suggesting shared pathogenic mechanisms.

These findings suggest that for patients with early-onset, familial, or levodopa-refractory parkinsonism—particularly after excluding secondary parkinsonism and negative genetic testing for hereditary PD—SCA8 screening should be considered alongside conventional SCA2/3/17 evaluations.

In this study, the parkinsonian phenotypes observed in SCAs predominantly manifested as akinetic-rigidity type rather than tremor-predominant type, with minimal cerebellar ataxia, consistenting with previous reports (20, 38). The emergence of parkinsonian features in SCAs may involve multiple determinants. First, SCA2, SCA3, and SCA17 cases presenting with parkinsonism share a common characteristic: shorter CAG repeat expansions (5, 12–14, 38). Prior studies indicate that SCA2 patients with parkinsonian phenotypes exhibit lower CAG repeats compared to those with ataxia-predominant presentations [36.2 ± 1.1 vs. 43.1 ± 3.2], alongside later symptom onset [45.8 ± 13.9 vs. 26.9 ± 11.0 years] (12). In our study, the proband 1(F1: II-4) with parkinsonism harbored 35 CAG repeats in ATXN2 and developed symptoms at age 37, aligning with these observations. Wu et al. identified repeat expansions at the SCA8 locus in 4/264 patients (1.5%) diagnosed with typical late-onset, levodopa-responsive Parkinson’s disease, with expansion sizes ranging from 75 to 92 repeats (20), and the SCA8 repeat numbers associated with parkinsonian phenotypes were relatively low. However, the limited cohort size precludes definitive conclusions regarding potential correlations between SCA8-associated parkinsonism and repeat expansion length, necessitating validation in larger cohorts. Second, ethnic disparities significantly influence phenotypic expression. For instance, parkinsonian phenotypes in SCA3 are more prevalent among individuals of African ancestry (2, 15), whereas SCA2-associated parkinsonism is more frequently observed in Asian populations (8, 16, 17). Current reports of SCA8 with parkinsonian manifestations are predominantly from Asian regions, including mainland China, Taiwan, South Korea, and Japan (20, 22–24, 36), these observations suggest that ethnic-specific factors may influence phenotypic expression in SCA8, and parkinsonism may be a common presentation of SCA8 in East Asian populations. Furthermore, sequence interruptions within repeat expansions may modulate phenotypic outcomes. CAA interruptions within CAG repeat expansions have been identified in SCA2 patients exhibiting Parkinson’s disease (PD)-like phenotypes, while absent in those without PD manifestations. These interruptions are postulated to stabilize repeat sequences during genetic transmission, though their precise mechanistic role in phenotypic divergence remains unclear (39). A parallel phenomenon of CAA interruptions has been observed in SCA17 patients with concurrent PD (13). SCA8 presents a distinct pattern where neither disease onset age nor severity correlates with pure repeat length. Instead, CCG•CGG interruptions appear to enhance the disease penetrance, with increasing numbers of interruptions inversely correlating with age of onset (40). This study did not assess CAG repeat interruptions in ATXN2 for proband 1 (F1: II-4). The SCA2-confirmed Family 1 (F1) exhibited intrafamilial phenotypic heterogeneity: while the proband manifested both parkinsonism and cerebellar ataxia, her mother and brother presented with pure cerebellar ataxia. Regrettably, genetic-phenotypic correlation analysis was precluded due to the mother’s death and brother’s refusal of genetic testing. In the SCA8 parkinsonism pedigree, similar CTA/CTG repeat expansions coexisted with divergent clinical features. As CCG•CGG interruption analysis was not performed, the potential contribution of such interruptions to this intrafamilial heterogeneity remains undetermined. Systematic analysis of interruption patterns will be prioritized in future mechanistic investigations.

Proband 5 (F5: II-4), who manifested Parkinson’s disease, was found to carry an SCA8 CTA/CTG expansion of 55 repeats. Studies on the pathogenic expansion threshold of ATXN8OS indicate that the CTA/CTG repeat numbers in most healthy individuals range from 15 to 50 in global populations. For affected individuals, the repeat length should be at least 50, while the number of pathogenic repeat length is more than 70 (41–43). However, studies in Chinese populations reveal distinct characteristics. Among 261 healthy controls, the CTA/CTG repeat numbers ranged from 12 to 43 (mean: 24.04 ± 4.53), with 18 repeats being the most frequent (41). These findings suggest an overall lower distribution of ATXN8OS CTA/CTG repeat numbers in the Chinese population compared to other ethnic groups. A symptomatic SCA8 case with 51 repeats manifesting cerebellar ataxia has been previously reported in China (44). In our case, the CTA/CTG repeat expansion was identified as 55 repeats. In contrast to previous case reports, the proband in this study presented with sporadic early-onset Parkinson’s disease without cerebellar ataxia and exhibited a positive response to levodopa therapy. We will longitudinally monitor this patient for potential emergence of cerebellar symptoms. This case demonstrates that ATXN8OS CTA/CTG repeat expansions within the range of 50 to 70 repeats can exhibit full disease penetrance.

Proband 3 (F3: III-3) developed Tourette syndrome comorbid with OCD during the preataxic stage of SCA3. To our knowledge, this represents the first documented case of Tourette syndrome in SCA3. SCAs are frequently associated with movement disorders (e.g., parkinsonism, choreiform movements, dystonia) (4), and non-motor comorbidities such as anxiety and depression, the latter being particularly prevalent in SCA3 (45). Tic Disorders (TD), a neuropsychiatric condition characterized by involuntary motor/vocal tics with childhood onset (46). According to previous studies, SCA3 typically exhibits inverse correlation between CAG repeat length and onset age (47, 48). Notably, in Family 3 (F3), affected members have similar CAG repeat lengths, and others developed ataxia around the age of 50. While proband 3 had an onset age much earlier than other affected members in the family. We hypothesize that proband 3 is most likely in the pre-ataxia stage, and Tourette syndrome is more likely a coincidental comorbidity that occurs independently of SCA3.

Proband 6 (F6: II-1) presented with an SCA17 subtype characterized by dystonia (torsional spasm) and spastic paraplegia, carrying a TBP gene CAG/CAA repeat expansion of 41 repeats without concurrent STUB1 heterozygous mutations. Previous studies define fully penetrant TBP alleles as CAG/CAA repeats ≥49 (49), while intermediate repeats (41–48) exhibit incomplete penetrance, where carriers may or may not develop symptoms. Federico et al. (50) reviewed 85 SCA17 cases with smaller CAG/CAA expansions (41–49 repeats), reporting a mean symptom onset age of 45 years (±13). Gait ataxia was the most common feature, followed by cognitive decline, parkinsonism, hyperkinetic disorders, and non-ataxic cerebellar signs (e.g., dysarthria). Their findings suggest that CAG/CAA repeats within 41–49 may still exert pathogenic effects (50). Magri et al. (51) proposed a digenic TBP/STUB1-associated SCA17 (SCA17-DI) mechanism, wherein co-occurrence of 41–46 CAG/CAA repeats and STUB1 pathogenic variants leads to complete phenotypic penetrance. The present case carried a CAG/CAA repeat expansion of 41 in the TBP gene and no STUB1 heterozygous mutations confirmed by NGS testing. This finding suggests that SCA17 with intermediate CAG/CAA repeat expansions (41 repeats) may achieve complete phenotypic penetrance even in the absence of coexisting STUB1 mutations. Therefore, in cases with clinical manifestations associated with intermediate TBP-expanded alleles reported in the literature, further screening for coexisting STUB1 heterozygous mutations is warranted to establish the minimum pathogenic repeat threshold for CAG/CAA expansions. Notably, previously reported cases with 41 CAG/CAA repeats predominantly manifested chorea with other movement disorders (52–56), whereas dystonia combined with spastic paraplegia, as observed here, has been infrequently documented. This case expands the clinical spectrum of SCA17 associated with 41 CAG/CAA repeat expansions.