Clinical Profile and Genetic Composition of Patients With Juvenile Parkinsonism From a Single Tertiary Care Center in India

Participants and clinical assessment

This retrospective chart review was conducted in patients with JP (onset ≤21 years) who were evaluated and managed at the National Institute of Mental Health and Neurosciences (NIMHANS), India, between 2015 and 2024. As a tertiary referral center, the hospital caters to the ethnically and geographically heterogeneous patient population across the country. Patients with JP of suspected genetic etiology who had undergone genetic testing were recruited for this study after other potential causes were carefully ruled out through clinical evaluation and appropriate investigations. The available demographic, clinical, and genetic data were collected from the hospital database, and appropriate clinical staging (modified Hoehn and Yahr [H&Y] staging and the Unified Parkinson’s Disease Rating Scale motor score-part III [UPDRS-III]) was documented.

Genetic testing

Exome sequencing was performed in all 40 patients to identify the underlying genetic etiology. In patients with negative next-generation sequencing results, copy number variant analysis of PD-associated genes was performed using either bioinformatic tools such as exome depth methods or multiple ligand probe assays (MLPAs).

Subgroup analysis

Patients were classified into the Gene (+) group when a pathogenic or a likely pathogenic variant was identified. Patients with a variant of uncertain significance (VUS) but with a consistent phenotype were also included in the Gene (+) subgroup. The remaining patients were classified into the Gene (-) subgroup.

Statistical analysis and ethics

Statistical analysis was performed using SPSS version 28.0 (IBM Corp.). Data are expressed in descriptive statistics as the means with standard deviations for continuous variables or as frequencies and percentages for categorical variables. The chi-square test for categorical variables and the Mann‒Whitney U test for continuous variables were performed to determine differences between the two groups. The NIMHANS ethics committee approved the study (No. NIMHANS/13/53rd IEC (BS&NS DIV.)/2025). Written informed consent was obtained from all the patients for video recording and publication. Sixteen patients were previously reported [12]. The study was partially funded by the Indian Council of Medical Research (ICMR), India (Project no.54/12/2019-HUM/BMS dated 30/09/2019).

Demographics and clinical characteristics

Forty patients (22 males, 55.0%) with JP of suspected genetic etiology were included in this study. Table 1 summarizes the key demographic and clinical characteristics of the entire cohort. The mean age at onset was 15.85±4.96 years, and the mean age at presentation was 26.37±10.11 years. The mean duration of illness at presentation was 10.43±10.49 years. The majority of the patients had an illness duration of less than 5 years (19 patients, 47.5%), whereas 14 patients (35.0%) had an illness duration >10 years, and 7 patients (17.5%) had an illness duration between 5 and 10 years. A positive family history was observed in 16 patients (40.0%), and consanguinity was present in 18 (45.0%).

Table 1.

Demographic, clinical, and treatment profile of the total cohort, gene positive and negative subgroup

Motor symptoms

Tremor was the most common symptom at onset (n=19, 47.5%), with isolated tremor in 13 patients (32.5%) and associated bradykinesia, rigidity, or dystonia in 6 patients (15.0%). Stiffness or rigidity was the first symptom in 12 patients (30.0%), and slowness or bradykinesia was the first symptom in 9 patients (22.5%). Five patients (12.5%) presented all 3 core parkinsonian symptoms from onset. Dystonia was the initial symptom in 10 patients (25.0%), whereas cognitive impairment was reported in 6 patients (15.0%) on the basis of clinical history at disease onset, and seizures were present in 3 patients (7.5%).

At presentation, slowness was the most common symptom, reported by 35 patients (87.5%), followed by tremor in 32 patients (80.0%) and stiffness in 30 patients (75.0%). Seven patients (17.5%) presented with gait difficulty and postural instability. Examinations revealed bradykinesia in 38 patients (95.0%), followed by rigidity and tremor in 33 patients (82.5%) each. Abnormal posturing or dystonia was the other major movement disorder and was observed in 30 patients (75.0%). More than three-fourths of the patients (88.6%) had a good levodopa response. Thirty percent of the patients eventually developed motor fluctuations, with peak-dose generalized choreiform dyskinesia in all of them (n=12) and wearing off in 7 patients (17.5%).

Nonmotor symptoms

The most common nonmotor symptom reported was cognitive impairment, which was seen in 7 patients (17.5%), followed by mood symptoms in 6 patients (15.0%). Among mood symptoms, depression was more commonly observed than anxiety was. Autonomic symptoms were reported by 5 patients (13.2%), all of whom reported urinary disturbances in the form of increased frequency, urgency, or urge incontinence, whereas only 3 patients reported constipation. Psychosis and rapid eye movement sleep behavior disorder (RBD) were present in 3 patients (7.9%) each. Three patients (7.9%) had a history of previous seizures as well.

Other neurological findings

Oculomotor abnormalities were observed in 21 patients (52.5%). Hypometric saccades were seen in 18 patients (42.1%), whereas mild vertical (up) gaze restriction and saccadic pursuits were observed in 12 patients (30.0%) each. Pyramidal signs were seen in 17 patients (42.5%), most commonly brisk deep tendon reflexes (n=16, 40%), followed by spasticity and extensor plantar responses in eight patients (20.0%) each.

Rating scales

The mean modified H&Y staging score was 2.62±0.88, with the majority of the patients being in stage 2 (n=21, 52.5%), followed by 7 patients (17.5%) each in stages 2.5 and 3. The mean UPDRS-III score (n=24) was 50.90±20.38 in the OFF state and 21.27±13.30 in the ON state, with a 54.38%±25.82% improvement. Approximately four-fifths of the patients (n=19, 79.2%) presented greater than 30% improvement in UPDRS-III scores. Mini-Mental State Examination (MMSE) scores were available for 14 patients, with a mean score of 22±7. Eight patients presented an MMSE score >23. The Montreal Cognitive Assessment (MoCA) score was available for only six patients, with a mean score of 11±7. Only one patient had a MOCA score >23.

Genetic architecture of the entire cohort

Twenty-seven patients (67.5%) had pathogenic/likely pathogenic variants or VUSs but with a consistent phenotype in either PD-associated or non-PD-associated genes, confirming the diagnosis of genetically determined JP (Figure 1). These 27 patients constituted the Gene (+) subgroup, and the remaining 13 patients with negative genetic tests constituted the Gene (-) subgroup. All positive cases involved recessively inherited genes, of which 23 patients presented a homozygous state, while the remaining 4 patients were in the compound-heterozygous state. In 22 patients, PD-associated genes were involved (PRKN [n=8], PLA2G6 [n=7], SYNJ1 [n=3], PARK7 [n=2], PINK1 [n=1], and ATP13A2 [n=1]), whereas in the remaining seven positive patients, non-PD-associated genes were involved (one case each involving PANK2, C19ORF12, SPG11, ATP7B, and GCH1).

Figure 1.

Genetic profile of juvenile parkinsonism. PD, Parkinson’s disease.

A total of 27 unique variants were identified in 11 different genes, of which 23 were pathogenic/likely pathogenic variants and the remaining four were VUSs according to the American College of Medical Genetics (ACMG) criteria. Missense variants were the most common (14 variants; PLA2G6: 4, SYNJ1: 3, PARK7: 1, ATP13A2: 1, PANK2: 1, C19ORF12: 1, GCH1: 2, and ATP7B: 1), followed by structural copy number variants (5 variants; all PRKN), stop-gain (4 variants; PRKN: 2, SYNJ1: 1, SPG11: 1), frameshift truncation (2 variants; PLA2G6: 1, PARK7: 1), splice-site (1 variant; PRKN: 1), and in-frame deletion variants (1 variant; PINK1: 1) (Table 2). The dot diagram in Figure 2 shows the distribution of various pathogenic genes according to the age at onset. Two patients with an age at onset less than 10 years had variants identified in the GCH1 and SPG11 genes. Compared with patients with C19ORF12, ATP7B, and PLA2G6, patients with PINK1, PARK7, SYNJ1, and PANK2 mutations had a later age of onset, approximately 20 years.

Table 2.

Details of the variants identified in the cohort

Figure 2.

Depicts the dot diagram showing the distribution of various pathogenic genes based on age of onset. MPAN, mitochondrial membrane protein-associated neurodegeneration; PKAN, pantothenate kinase-associated neurodegeneration.

Neuroimaging findings

In our cohort of 35 patients whose brain magnetic resonance imaging (MRI) scans were available, several gene-associated patterns were observed. Among eight individuals with PARK-PRKN, imaging was available for five: two had normal MRI scans, while three had subtle globus pallidus mineralization, and the loss of the “swallow tail” sign was observed in one patient. All patients with PARK-PLA2G6 presented with abnormal findings, such as cerebellar atrophy, claval hypertrophy, vertical orientation of the splenium of the corpus callosum (in two cases), cerebral atrophy (in two cases), and mineralization in the bilateral globus pallidus, substantia nigra, and lentiform and caudate nuclei. A patient with mitochondrial membrane protein-associated neurodegeneration (MPAN or NBIA-C19orf12) had blooming in the basal ganglia on susceptibility-weighted imaging, suggesting mineralization. Among the three patients with PARK-SYNJ1, two had normal imaging results, and the third had mild cerebral and cerebellar atrophy. Patients with Wilson’s disease presented classical bilateral putaminal T2 and fluid attenuated inversion recovery hyperintensities with diffusion restriction. Patients with variants in PINK1, PARK7, ATP13A2, GCH1, and SPG11 had normal or nonspecific imaging findings.

Among the Gene (-) patients, most had normal or atrophic brains; one showed bilateral globus pallidi interna mineralization. Unfortunately, dopamine transporter (DAT) scan results were not available for the majority of our cohort; therefore, a systematic analysis could not be performed.

Treatment and follow-up

Thirty-five patients (87.5%) were receiving levodopa-carbidopa, with a mean dose of 289±197 mg. Fourteen patients were receiving a dopamine agonist: pramipexole in 12 patients (30%), with a mean dose of 1.50±0.74 mg, and ropinirole in 2 patients (5.0%). Trihexyphenidyl was given to 8 patients (20.0%) who had dystonia, and amantadine was given to 6 patients (15.8%) with motor fluctuations. The mean levodopa equivalent daily dose (LEDD) was 309.27±275.80 mg. Two patients underwent bilateral subthalamic nuclei deep brain stimulation (DBS). One patient with a PARK-PRKN showed significant clinical improvement at 5 years post-DBS, whereas another patient from the Gene (-) group developed clinical deterioration after an initial response. The follow-up details of 18 patients were available, with a mean duration of illness of 10.1±9.5 years. There were no significant side effects associated with medications except for motor fluctuations at the last follow-up. Patients who had undergone DBS showed some initial improvement but developed clinical worsening after 1–2 years.

Subgroup analysis

Clinical features of patients with PARK-PRKN and PARK-PLA2G6

PARK-PRKN and PARK-PLA2G6 were the two most common disease subgroups affecting 8 and 7 patients, respectively. Rigidity and bradykinesia were seen in the patients, whereas tremors were more common in the PARK-PRKN group. Dystonia (p=0.070), up gaze restriction (p=0.067), and pyramidal signs (p=0.019) were more common in PARK-PLA2G6 patients, and they had more severe illness at presentation, as indicated by an H&Y score ≥3 (p=0.057). Moreover, the mean LEDD was greater (p=0.043) in the PARK-PRKN group, with more motor fluctuations (p=0.067) (Table 3). PARK-PLA2G6 patients also had a younger age at onset, younger age at presentation, and shorter duration of illness (Figure 3). Nonmotor symptoms were more frequent, with a higher mean UPDRS-III score and a weaker levodopa response. These observations were not statistically significant.

Table 3.

Demographic, clinical, and treatment profile of the PARK-PRKN and PARK-PLA2G6 subgroups

Figure 3.

Comparison of the PARK-PRKN and PARK-PLA2G6 cohorts. RBD, rapid eye movement sleep behavior disorder; UPDRS-III, Unified Parkinson’s Disease Rating Scale motor score-part III; H&Y, Hoehn and Yahr.

Approach to JP of suspected genetic etiology

JP can be attributed to abnormalities in genes traditionally associated with PD and genes linked to other neurodegenerative or metabolic disorders [15]. PD-associated genes involved in JP are broadly categorized into two groups: genes causing typical, idiopathic-like PD and genes associated with atypical or complex parkinsonian syndromes. Genes such as PRKN, PINK1, and PARK7 are commonly related to early-onset, levodopa-responsive Parkinsonism, resembling classical PD but with early motor fluctuations [8]. In contrast, abnormalities in the PLA2G6, ATP13A2, SYNJ1, DNAJC6, VPS13C, or FBXO7 genes are associated with atypical presentations, often involving additional neurological features such as early cognitive decline, behavioral abnormalities, seizures, suboptimal levodopa response, ocular abnormalities, prominent dystonia, pyramidal signs, and cerebellar involvement [6,8].

Parkinsonian syndromes with onset within the first year of life are also recognized and are referred to as infantile parkinsonism-dystonia. These conditions are typically caused by autosomal recessive, severe inborn errors of dopamine metabolism, such as those involving tyrosine hydroxylase (TH), aromatic L-amino acid decarboxylase (AADC), or DAT, leading to profound dopamine deficiency [16]. An illustration of the genetic etiology of JP is shown in Figure 4.

Figure 4.

Depicts the approach to genetic etiologies of juvenile parkinsonism. PD, Parkinson’s disease; AD, autosomal dominant; AR, autosomal recessive; HSP, hereditary spastic paraplegia; NBIA, neurodegeneration and brain iron accumulation; SCA, spinocerebellar ataxia.

The initial assessment of a patient with JP should focus on ruling out acquired and treatable causes, such as drug- or toxin-induced cases and immune-mediated or infection-related cases. Neuroimaging, toxin/drug screening, and immune workup, including an autoimmune encephalitis panel, are needed on the basis of clinical suspicion. Screening for Wilson’s disease is essential in all patients before genetic testing and should include serum copper, serum ceruloplasmin, and 24-hour urinary copper levels [6,17].

Genetic testing is indicated after ruling out acquired causes or early in the evaluation if there is a strong suspicion, as indicated by a positive family history or clinical features that indicate a particular genotype [3]. Targeted genetic testing could be advised when a phenotype strongly suggests a specific genetic etiology. Broader genetic approaches, such as parkinsonism gene panels or whole-exome sequencing, may offer higher diagnostic yield without clear clinical clues. These methods are especially useful in cases with atypical presentations or negative targeted testing results. Regardless of etiology, all patients with significant symptoms should be offered symptomatic treatment tailored to their clinical profile to improve their quality of life.

Levodopa-carbidopa should be considered in all patients with parkinsonism and dystonia, although early levodopa-induced dyskinesias and motor fluctuations are common [18]. In cases of dose-limiting side effects, standard adult strategies may be employed. Dopamine agonists are less effective but can be given [19]. Anticholinergics, such as trihexyphenidyl, are generally well tolerated in children and are indicated for the treatment of dystonia and levodopa-refractory tremors. Botulinum toxin and deep brain stimulation have demonstrated safety and efficacy in pediatric populations, particularly for managing focal dystonia and generalized dystonia parkinsonism, respectively [20,21].

Review of the literature on JP

The literature on the clinical profile of JP is limited, especially for the Asia–Oceania region. Earlier studies have reported a high frequency of PRKN gene variants in JP. In a study by Zagorovskaia et al. [22], clinical and genetic analysis of 26 siblings from 20 families revealed PRKN variants in 41% of cases, which was higher than the frequencies reported in studies by Lücking et al. [23] and Periquet et al. [24]. A recent review analyzing clinical, imaging, and genetic data from 407 JP patients across 219 studies revealed 60 distinct genetic disorders [25]. Among 404 patients whose data were reported, the median age of symptom onset was 10 years (range: 0–21 years), which was nearly a decade earlier than that in our cohort (18 years [12.75–20.00 years]). Resting tremor was observed in only 26.7% (n=109 patients), and data on rigidity and bradykinesia were not available. Dystonia, pyramidal signs, and abnormal eye movements were less common in our cohort (61%, 35.3% and 27.2% vs. 75%, 42.5%, and 52.5%, respectively), whereas cognitive impairment, neuropsychiatric symptoms, and epilepsy were more common (54.8%, 28.5%, and 12% vs. 17.5%, 15%, and 7.9%, respectively). Other common associated features included ataxia (17.9%), myoclonus (13.8%), chorea (8.6%), and neuropathy (8.0%), which were not observed in our patients. Levodopa response was reported in 250 individuals (61.4%), which was less frequently than in our patients (88.6%) [25].

According to a review of the literature, the most frequent genotypes identified in decreasing order of frequency were PRKN, HTT, ATP13A2, ATP1A3, FBX07, PINK1, and PLA2G6. The most common variants were in the PRKN gene, identified in 179 patients (43.95%), who were found to have typical PD features, with a median age at onset of 17 years [25]. In contrast, up to 92% of patients with the other genetic subtypes had atypical presentations. PLA2G6 variants were identified in 16 patients with predominantly atypical presentations, including cognitive decline, oculomotor abnormalities, neuropsychiatric manifestations, and pyramidal signs, similar to our patients [25].

Review of the Indian literature on JP

In a review of the Indian literature on JP, we identified only 15 publications, of which clinical details and genetic etiology were available in 6. Five were single case reports of variants in the SYNJ1, PODXL, DNAJC6, PARK7, and BTD genes [26-30]. A comparison between juvenile- and young-onset PD by Muthane et al. [31] in 1994 and a meta-analysis of previously published cases revealed that patients with JP had a greater familial occurrence of PD and dystonia; moreover, autonomic symptoms were twice as common in patients with JP, and all patients with JP showed a meaningful response to levodopa. In the largest genetic study of early-onset PD from the South Asian population to date, 23 of the 674 patients had juvenile-onset PD [14]. The age at onset was slightly greater in our cohort (16.50±3.85 years). Compared with the current study, there was less family history (30.4% vs. 40.0%) and more consanguinity (56.5% vs. 45.0%). As in our patients, bradykinesia was the most common symptom (91.3%), followed by tremor and rigidity (78.3%). Postural and gait impairments were observed more commonly in our cohort (34.8% and 75.0%, respectively). The frequency of nonmotor symptoms was slightly greater than that in our cohort. The genetic yield was 43.47% (10/23), which was less than that of the current cohort. The most common genetic etiology was PARK-PRKN (n=7), followed by PARK-PLA2G6 (n=2) and PARK-ATP13A2 (n=1) [14].

In a recent study of 16 patients with PRKN variants from our institution, the median age at onset was 28.5 years [32]. Parkinsonism (93.8%) and dystonia (62.5%) were the most common motor features, with tremors being the most frequent initial symptoms. Nonmotor symptoms were present in 62.5% of the patients, and all the patients responded well to levodopa, although 50% of the patients developed drug-induced dyskinesia. In another study of 26 patients with PLA2G6 variants, the median age at onset was 13.0 years [33]. Dystonia (53.8%) and parkinsonism (42.3%) were the predominant motor features. Nonmotor symptoms such as cognitive decline (46.2%) and behavioral changes (23.1%) were notable. Levodopa was effective in most patients, but dyskinesias were common.

The results of the current study revealed a genetic yield of 67.5%, which is higher than that in previous publications, probably because the referral center is a tertiary referral center [9]. The Gene (+) patients were younger and had more severe disease at presentation, with a better levodopa response and more frequent cognitive decline, oculomotor abnormalities, pyramidal signs, motor fluctuations, and dyskinesia.

All the commonly identified PD-associated genes were autosomal recessive, and the majority had a typical phenotype. PRKN and PLA2G6 were the most common genes, and compared with PARK-PRKN patients, PARK-PLA2G6 patients were more likely to have nonmotor symptoms, severe disease, a weaker levodopa response, and dyskinesia. A comparison between the first decade and the second decade of onset was attempted but could not be performed because of a significant disparity in the number of cases. Only two genes were identified in patients whose age of onset was less than 10 years; both were not associated with PD, namely, GCH1 and SPG11. Patients with PINK1, PARK7, and PANK2 variants had a later onset, toward the end of the second decade.

Limitations

Although our study represents one of the largest single-center cohorts of JP patients to date, it has certain limitations because of its retrospective design. Comprehensive phenotyping was impossible, as many symptoms, particularly nonmotor symptoms, were not assessed using validated clinical scales. Consequently, symptom severity was assessed on the basis of the presence or absence of features documented in the clinical history and examination. Additionally, the clinical progression of individual symptoms was not consistently documented, and follow-up data were available for only a limited number of patients. Exome sequencing and MLPA may miss certain genetic abnormalities, such as repeat expansions or deep intronic variants. For instance, juvenile-onset spinocerebellar ataxias and Huntington’s disease (HD) can present as JP and require a triplet-primed polymerase chain reaction method for identification. However, none of the Gene (-) patients had any autosomal dominant inheritance, suggesting the possibility of spinocerebellar ataxic or juvenile HD presentations. Complex structural variants are being increasingly recognized in the PRKN gene, which can be negative even on MLPA and would require long-read sequencing for its identification [34]. Another possibility is that deep intronic variants cause aberrant splicing, thereby resulting in disease. Whole-genome sequencing will help in such cases, or if there is a strong clinical suspicion of a specific genetic etiology, then messenger ribonucleic acid sequencing would also help in such cases [35]. These mechanisms may explain some of the Gene (-) cases in our cohort. Furthermore, as a tertiary referral center, our cohort may not fully represent the general population, as more complex and atypical cases are more likely to be referred to us.

Conclusion

This study reports the genetic profile of the first Indian cohort of JP patients. More than two-thirds (67.5%) of the JP patients in our cohort had an underlying monogenic cause, with PARK-PRKN, PARK-PLA2G6, and PARK-SYNJ1 being the common causes of genetically mediated JP in India. Gene (+) patients had greater occurrences of familial consanguinity, oculomotor abnormalities, motor fluctuations, and dyskinesia; more severe disease; and better levodopa responsiveness at presentation. Dystonia, gaze restriction, and pyramidal signs were more common in PARK-PLA2G6 patients, whereas motor fluctuations could be seen more frequently in PARK-PRKN patients. More indepth research is essential for enhancing our understanding and comprehensively characterizing the genetic spectrum of JP.