Utility of Clinical Exome Sequencing in Dystonia: A Single-Center Study From India.

OBJECTIVE: With the use of next-generation sequencing in clinical practice, several genetic etiologies of dystonia have been identified. This study aimed to ascertain the utility of clinical exome sequencing (CES) in dystonia and factors suggestive of a genetic etiology.
METHODS: This study was a retrospective chart review of patients with dystonia who had undergone CES for the evaluation of dystonia.
RESULTS: Forty-eight patients (35 males, 46 families) with dystonia were studied, with a mean age at onset of 16.0 ± 14.1 (1-58) years. A pathogenic/likely pathogenic variant was found in 20 patients (41.7%) among which 14 patients (29.2%) carried a novel variant. CES was more likely to detect a genetic diagnosis in patients with an early age at onset, i.e., ≤ 20 years.
CONCLUSION: CES is a useful tool in the diagnostic evaluation of dystonia, with a yield of close to 40%. Patients with an earlier age at onset have a higher likelihood of having dystonia due to a genetic cause than those with a later age at onset.

Dystonia is a common hyperkinetic movement disorder characterized by sustained or intermittent muscle contractions causing abnormal, often repetitive, movements or postures or both [1]. Clinical details such as family history, onset age, dystonia pattern, associated abnormal movements, and neurological and systemic involvement aid in formulating a diagnosis of dystonia [2]. Imaging and genetic assessments provide additional information that is often sufficient to confirm the diagnosis. However, owing to the numerous, complex and undescribed etiologies, a diagnosis may not be possible in every patient.

Genetic assessments are particularly helpful in the identification of monogenic forms, but individual gene testing is tedious and time-consuming due to the phenotypic heterogeneity of single-gene disorders and the phenotypic overlap of many different genetic disorders. Next-generation sequencing (NGS) technology allows quick screening of thousands of genes and the simultaneous identification of known and novel variants. This technology has expedited the process of confirming genetic diagnoses and the discovery of novel pathogenic genes and variants.

In a clinical setting, NGS can be used to evaluate dystonia through various panels, such as targeted gene panels, clinical exome sequencing (CES), whole-exome sequencing (WES), or whole-genome sequencing (WGS), in the order of increasing coverage, diagnostic yield, interpretation complexity, and cost. As the name suggests, WGS and WES involve sequencing the whole genome and whole exome of over 21,000 genes identified so far, respectively. However, CES involves the sequencing of an exome subset of approximately 3,000–6,000 genes chosen based on association with the disease, and the subset may vary depending on the technology provider and with time based on the newly available literature. The diagnostic yield varies from 1% to 37.5% among different studies based on the subjects studied and the techniques used [3-10]. In this study, we retrospectively reviewed the CES results of patients presenting with dystonia as the predominant symptom who had undergone CES as part of their diagnostic evaluation to assess the utility of CES and ascertain factors suggestive of a positive genetic diagnosis.

MATERIALS & METHODS

In this retrospective chart review, we reviewed our movement disorders database from 2016 to 2020 to identify those patients who had undergone CES as part of the evaluation of dystonia as their prominent phenotypic feature. Patient details were anonymized to maintain patient privacy. The clinical data, investigations and genetic reports of these patients were reviewed. All these patients were examined by movement disorder specialists PKP, RY, NK, VVH). Dystonia in each patient was classified according to the recent consensus update on the classification and phenomenology of dystonia [1]. Additionally, based on the associated neurological features, dystonia was classified as isolated dystonia (IsoD), combined dystonia (CombD) or complex dystonia (CxD). Patients with IsoD had dystonia as the predominant feature, whereas patients with CombD had dystonia with other movement disorders, such as myoclonus, chorea, or parkinsonism. Patients with CxD had other neurological or systemic features [11]. CES (Supplementary Materials 1 and 2 in the online-only Data Supplement) was performed in all these patients, and the report was analyzed. The Institute Ethics Committee at the National Institute of Mental Health and Neurosciences granted an ethical clearance waiver owing to the retrospective nature of the study, with deidentified data being extracted from the files (No: NIMH/DO/DEAN [Basic Science]/2020-21). p value of < 0.05 considered statistically significant (Supplementary Material 1 in the online-only Data Supplement).

RESULTS

Forty-eight patients (35 males) belonging to 46 families were identified (Table 1). None of these patients had undergone single-gene or targeted-gene testing prior to CES. A positive family history was noted in 7 patients, including in 2 patients each from 2 families and in one patient each from 3 families (5 total families). Three patients from our center whose cases were previously published (patient 5 [12], patient 12 [13], and patient 16 [14]) were included in this study. The demographic and clinical details and axis-1 classification details of this cohort are provided in Table 1.

Table 1.

Overall demographics and clinical manifestations of probands with dystonia

Features	Overall (n = 48)	Genetic diagnosis		p value
		Yes (n = 20)	No (n = 28)
Sex (male/female)	35/13	15/5	20/8	0.78
Age at testing (years)	25.2 ± 14.0	22.5 ± 11.2	27.0 ± 16.1	0.28
Age at onset (years)	16.0 ± 14.1	11.9 ± 9.6	19.1 ± 16.3	0.08
Infantile onset	3 (6.2)	2 (10)	1 (3.5)	0.56
Childhood onset	25 (52)	11 (55)	14 (50)	0.77
Adolescence	8 (16.6)	6 (30)	2 (7.1)	0.05
Early adulthood	5 (10.4)	0	5 (17.8)	-
Late adulthood	7 (14.6)	1 (5)	6 (21.4)	0.21
Body distribution				-
Focal dystonia	7 (14.6)	1 (5)	6 (21.4)	0.21
Multifocal dystonia	4 (8.3)	4 (20)	0	-
Segmental dystonia	4 (8.3)	1 (5)	3 (10.7)	0.63
Generalized dystonia	33 (68.7)	14 (70)	19 (67.8)	> 0.99
Associated features				-
Isolated dystonia	18 (37.5)	6 (30)	12 (42.8)	0.54
Combined dystonia	18 (37.5)	9 (45)	9 (32.1)	0.4
Dystonia-parkinsonism	13 (72.2)	7 (77.8)	6 (66.7)	> 0.99
Dystonia-chorea	4 (22.2)	2 (22.2)	2 (22.2)	> 0.99
Dystonia-myoclonus	1 (5.6)	0 (0)	1 (11.1)	-
Complex dystonia	12 (25.0)	5 (25)	7 (25)	> 0.99
Dystonia-cerebellar ataxia	4 (33.3)	1 (20)	3 (42.8)	0.60
Dystonia-spasticity	5 (41.7)	1 (20)	4 (57.1)	0.30
Dystonia with cognitive decline	2 (16.7)	2 (40)	0	-
Dystonia with intellectual disability	1 (8.3)	1 (20)	0	-
Family history (families)	5/46 (10.9)	3 (16.6)	2 (7.1)	0.40

Data are presented as mean ± standard deviation or number (%).

Pathogenic/likely pathogenic (P/LP) variants were present in 20 patients (18 families) (Table 2). The genes in which the P/LP variants were observed were PLA2G6 (4 patients from 3 families); GLB1 (3 patients from 2 families); TH, PRKN, TOR1A, and GCH1 in 2 patients each (a total of 8 patients from 8 families); and THAP1, NDUFA12, NPC1, ATP7B, and FA2H in one patient each (a total of 5 patients from 5 families) (Table 2). Among these, 14 patients (from 13 families) had novel variants that were classified as P/LP based on American College of Medical Genetics and Genomics (ACMG) causality criteria (Table 2). The age at onset (AAO), distribution of dystonia, associated features, and genetic abnormality in each patient are mentioned in Table 2. Among patients with a positive genetic diagnosis, TOR1A causing DYT1 was the most common genetic abnormality among those with IsoD, seen in 2 out of 6 patients, while PLA2G6 causing PLA2G6-associated neurodegeneration (PLAN) was the most common genetic abnormality among those with CombD, observed in 4 out of 9 patients. In addition, 5 patients had P/LP variants in the heterozygous carrier state, and 8 patients had variants of unknown significance (VUSs) in genes relevant to dystonia (Supplementary Table 1 in the online-only Data Supplement).

Table 2.

Genetic causes identified in the present study

Case	AAO	Phenotype	Gene	Variant	Inheritance	ACMG classification
1	15	Focal, combined (DP)	PLA2G6	ENST00000332509;c.835delA;p.Ile279SerfsTer26[†]	AR, CH	LP (PVS1PM2)
				ENST00000332509;c.991G>T;p.Asp331Tyr[†]		LP (PM2PP2-5)
2	10	Generalised, combined (DP)	PLA2G6	ENST00000332509;c.2222G>A;p.arg742Gln	AR, Hm	P (PP5PM2,5PP2)
3[*]	18	Generalised, combined (DP)	PLA2G6	ENST00000332509;c.2222G>A;p.arg742Gln	AR, Hm	P (PP5PM2,5PP1,2)
4[*]	19	Generalised, combined (DP)	PLA2G6	ENST00000332509;c.2222G>A;p.arg742Gln	AR, Hm	P (PP5PM2,5PP1,2)
5	7	Generalised, isolated	GLB1	ENST00000307363;c.246-2A>G	AR, CH	P (PVS1PM2PP3)
				ENST00000307363;c.1325G>A;p.Arg442Gln		P (PP5PM2,3PP2,3)
6[*]	4	Generalised, combined (DP)	GLB1	ENST00000307363;c.1325G>A;p.Arg442Gln	AR, CH	P (PP5PM2PP1-3)
				ENST00000307363;c.1022G>T;p.Gly341Val[†]		LP (PM2,3PP1-3)
7[*]	3	Generalised, combined (DP)	GLB1	ENST00000307363;c.1325G>A;p.Arg442Gln	AR, CH	P (PP5PM2PP1-3)
				ENST00000307363;c.1022G>T;p.Gly341Val[†]		LP (PM2,3PP1-3)
8	1	Multifocal, combined (DCh)	GCH1	ENST00000491895;c.542T>C;p.Val181Ala[†]	AR, Hm	LP (PM1,2PP2,3)
9	1	Generalised, complex (DId)	GCH1	ENST00000491895;c.614T>C;p.Val205Ala[†]	AR, CH	LP (PM1,2,5PP2,3)
				ENST00000491895;c.610G>A;p.Val204Ile		LP (PM1,2PP2,3)
10	11	Generalised, isolated	TOR1A	ENST00000351698;c.907_909del;p.Glu303del	AD, Ht	P (PP5PS3PM1,2,4PP3)
11	8	Multifocal, isolated	TOR1A	ENST00000351698;c.907_909del;p.Glu303del	AD, Ht	P (PP5PS3PM1,2,4PP3)
12	20	Generalised, isolated	TH	ENST00000381178;c.525delG;p.Leu176SerfsTer61[†]	AR, CH	LP (PVS1PM2)
				ENST00000381178:c.1481C>T;p.Thr494Met[†]		LP (PM2PP2,3,5)
13	17	Multifocal, combined (DCh)	TH	ENST00000381178;c.1117C>T;p.Arg373Cys[†]	AR, Hm	LP (PM1,2PP2,3)
14	20	Generalised, isolated	PRKN	ENST00000366898;c.124C>T;p.Arg42Cys[†]	AR, CH	LP (PM1,2,5PP2,3)
				ENST00000366898;c.1076G>A;p.Gly359Asp[†]		LP (PM1,2PP2,3)
15	17	Segmental, combined (DP)	PRKN	ENST00000366898;c.(171+1_172-1)_(412+1_413-1)del[†]	AR, Hm
16	43	Multifocal, isolated	THAP1	ENST00000254250;c.71+1G>C[†]	AD, Ht	LPVS1PM2PP3)
17	6	Generalised, complex (DS)	NDUFA12	ENST00000327772;c.60_61delCCinsT;p.Arg21GlufsTer18[†]	AR, Hm	LP (PVS1PM2)
18	13	Generalised, complex (DCo)	ATP7B	ENST00000242839;c.3485C>T;p.Ser1162Phe[†]	AR, Hm	P (PS3PM1,2PP2-4)
19	10	Generalised, complex (DC)	FA2H	ENST00000219368;c.130C>A;p.Pro44Thr[†]	AR, Hm	LP (PM1,2,5)
20	2	Generalised, complex (DCo)	NPC1	ENST00000269228;c.2473T>C;p.Tyr825His[†]	AR, Hm	LP (PM2,5PP2,3)

belongs to same family and segregation analysis was performed by Sanger analysis;

novel mutations.

AAO, age at onset; ACMG, American College of Medical Genetics and Genomics; AD, autosomal dominant; AR, autosomal recessive; CH, compound heterozygous; DC, dystonia cerebellar; DCh, dystonia chorea; DCo, dystonia with cognitive decline; DId, dystonia with intellectual disability; DP, dystonia parkinsonism; DS, dystonia spasticity; Hm, homozygous; Ht, heterozygous; LP, likely pathogenic; P, pathogenic; PVS, very strong evidence of pathogenicity; PS, strong evidence of pathogenicity; PM, moderate evidence of pathogenicity; PP, supportive evidence of pathogenicity.

When comparing patients with a positive genetic diagnosis to those with a negative genetic diagnosis, most of the patients with a positive genetic diagnosis had an early AAO of dystonia (≤ 20 yrs) (Table 1). Of an overall 36 patients with early-onset dystonia (≤ 20 years), 19 had P/LP mutations (52.8%). Among 12 patients with an adult onset of dystonia (> 20 years), only one had a P/LP mutation (8.3%) (p = 0.007). There was no significant difference in the diagnostic yield of genetic testing based on the distribution or the associated features.

DISCUSSION

In the present study, a cohort of 48 patients with heterogeneous dystonia who had undergone CES as part of their evaluation was analyzed. Our cohort had an early AAO of dystonia, with three-fourths of the patients having an AAO before 20 years. In comparison, earlier studies reported a relatively equal distribution of the AAO (Supplementary Table 2 in the online-only Data Supplement) [5,6,8-10]. Dystonia with an underlying genetic cause often has an early AAO and presentation. The yield of genetic testing in our cohort was higher (39.1%) than that in previous studies (11.7% to 37.5%) (Supplementary Table 2 in the online-only Data Supplement). This variability in diagnostic yield is most likely attributable to the inclusion criteria and possibly due to the differences in the population studied and the clinical characteristics of the enrolled patients. The higher yield seen in the present study may be due to the higher prevalence of young-onset dystonia and to selection bias. In previous studies, a higher yield of approximately 35% has been reported when the mean age of onset is in the 1st to 2nd decade [5,6,10]. The lower positive rate in some of these studies could be due to the inclusion of patients with negative results in the initial targeted gene testing of a single gene or a set of genes [6,9].

Even though genetic testing plays a crucial role in the diagnosis of an underlying genetic abnormality, it may not always be accessible or financially feasible. An awareness of the predictors for positive genetic testing would help in the judicious use of NGS. The comparison of patients with and without a positive genetic test revealed the importance of an AAO of ≤ 20 years. None of the remaining clinical features had any impact on the outcome of genetic testing. Previous studies have shown that patients with an early AAO CombD or CxD have a higher likelihood of having a positive genetic diagnosis. However, patients with a later AAO isolated or focal dystonia have a lower likelihood of a positive genetic diagnosis [6,9,10]. However, these findings are inconsistent. The lack of such associations in the present study despite the presence of patients with IsoD, CombD, and CxD may be due to a lower sample size or different genetic population makeup than in previous studies.

The spectrum of affected genes seen in the present study was different from that in previous studies (Supplementary Table 2 in the online-only Data Supplement) [3,5,6,8-10]. In the present study, PLA2G6 mutations were identified in 3 families (4 patients) with P/LP mutations (one novel mutation) and 2 families (2 patients) with VUSs. These patients presented with the typical dystonia-parkinsonism phenotypic subtype of PLAN. In another study from India [15], 3 patients had dystonia among 12 with genetically confirmed PLAN with a phenotype of neuronal axonal degeneration, suggesting that PLA2G6 mutations may not be an uncommon genetic cause in the Indian population. However, this may be restricted to the population studied, and larger studies are required to ascertain the prevalence of PLA2G6 mutations in patients with dystonia worldwide. DYT-TOR1A is considered to be the most common early-onset isolated genetic dystonia, which was the case in our study, and one-third of patients with IsoD had a previously reported pathogenic variant (p.Glu303del) in the TOR1A gene [9]. Previous studies from India vary in this regard. None of the patients among a cohort of 178 IsoD patients harbored the p.Glu303del variant in TOR1A [16], while in a different study, only two patients belonging to a single family in a cohort of 321 patients with IsoD harbored this mutation [17]. In the current study, a patient who presented with childhood-onset complex generalized dystonia was found to have an NDUFA12 gene mutation leading to Leigh’s syndrome. Although this is a known manifestation, it has not been described in previous studies of this nature. We did not observe any patients with ANO3, GNAL, or KMT2B mutations, which are among the recently described monogenic DYTs. This is consistent with previous reports with only a single patient with dystonia each due to GNAL [18] and KMT2B [19] mutations described thus far from India and none due to ANO3 mutations. The frequency of a particular genetic mutation depends on population studies, which also has to be kept in mind when designing targeted gene panels to suit the population being tested.

With the introduction of NGS, the number of variants identified has increased. We found VUSs in nearly as many patients as those who had a P/LP mutation. With advancements in genomics, the status of these VUSs will be clarified, which may further improve detection rates. However, one needs to be aware of the benefits and limitations of the tests being ordered. Although NGS is time-saving and cost effective, many VUSs can be detected, which need to be interpreted with caution to determine pathogenicity. Additionally, with WES and WGS, VUSs unrelated to the disease being evaluated may be detected, further compounding the difficulty in interpretation and adding to the psychological burden of patients owing to the presence of an additional unexpected abnormality. This possibility has to be discussed during pregenetic test counseling. Furthermore, a negative NGS result does not rule out a genetic disease, as NGS may miss copy number variants, large deletions and duplications, repeat disorders and diseases with imprinting phenomena. A genetic diagnosis has many clinical implications. Apart from identifying treatable disorders, an early genetic diagnosis will lead to proper prognostication, genetic counseling, family planning, and predicting the response to specific management options such as deep brain stimulation, which may have a differential response depending upon the genetic diagnosis [20].

Limitations

This study has several limitations owing to its retrospective nature. Not all patients with dystonia were offered or underwent genetic assessment. In addition, patients were not chosen consecutively or in any predefined manner, leading to a bias in ordering the genetic testing. The prevalence of patients with a younger AAO may be due to a clinical bias when ordering CES. It is likely that younger patients were advised CES more often in the hope of identifying a treatable cause. Biochemical estimations were not performed in patients with underlying enzyme defects, such as dopa-responsive dystonia, GM1 gangliosides, and Niemann-Pick disease type C. Segregation and functional studies could not be performed in several patients with VUSs to identify pathogenicity.

Conclusion

CES is a very useful tool in the evaluation of dystonia, with a yield of close to 40%. In particular, phenotypic and genotypic heterogeneity preclude single-gene tests, as they are time-consuming and not cost effective. An early AAO of dystonia could suggest a higher probability of a genetic cause and needs to be considered when ordering genetic testing.