Genetic Testing of Movements Disorders: A Review of Clinical Utility

Diagnostic Yield

The simplest and most quantifiable metric of diagnostic utility is diagnostic yield, the proportion of genetic diagnoses established for a given cohort using a particular genetic testing approach. Diagnostic yield is dependent on multiple factors, including but not limited to, the characteristics of the target population (e.g., type of movement disorder, presence/absence of family history, age at symptom onset, ethnic background), the type of genetic testing used (e.g., Sanger sequencing, short-read NGS, long-read sequencing, repeat-expansion testing using Southern blot and/or repeat-primed polymerase chain reaction [RP-PCR]), the sequencing target (e.g., single gene, targeted gene panel, whole exome sequencing [WES] or whole genome sequencing [WGS]), whether singleton, trio and/or family data is available for analysis, the type and extent of bioinformatic analysis and how strictly standardized variant curation criteria are applied to the classification of variants [8].

A systematic review of the diagnostic yield of genetic testing in all reported movement disorder cohorts is beyond the scope of this article. Instead, we have summarized the diagnostic yields of a non-exhaustive selection of recent (2021 onwards) studies that used prospectively and uniformly applied, genetic testing approaches in composite movement disorder cohorts, as well as cohorts of three selected common movement disorders; ataxia, dystonia and Parkinson’s disease (PD). The rationale for selection of only recent studies is the rapidly increasing number of genetic loci associated with movement disorders, which makes comparison of studies conducted even several years apart difficult; earlier studies have been summarized elsewhere [9, 10]. Retrospective cohort studies were excluded to minimize the potential influence of selection bias and heterogeneously applied genetic testing algorithms on diagnostic yield.

Despite restriction to recent prospectively tested cohorts, comparison between the studies presented in Table 1 is still limited by the heterogeneous nature of the studies, including differing methods for patient selection (e.g., inclusion/exclusion criteria) and contexts in which the genetic testing was performed (e.g., genetic testing on a clinical basis vs. research basis). Accepting these limitations, the overall diagnostic yield from the studies for composite movement disorder cohorts was 21–32%, giving an overall estimate of the diagnostic yield capacity of current genetic testing technologies for movement disorders. However, given that most of these studies recruited patients from large tertiary referral centers often with highly selected patient cohorts and some studies analyzed genetic testing that was performed on a research basis, it may be difficult to estimate the yield of these genetic testing strategies when applied on a clinical basis to real world movement disorder patients in the general community.

Of the three specific movement disorder cohorts considered in Table 1, diagnostic yield was highest for ataxia (33–55%), moderate for dystonia (12–21%) and lowest for PD (9–13% in unselected cohorts but increasing up to 26% in enriched subpopulations consisting of early-onset and familial PD). These differing yields could reflect intrinsic differences in the genetic architecture of these disorders and/or differential contributions of non-genetic factors (e.g., environmental, epigenetic) to these disorders [3, 11, 12]. For each disorder, various phenotypic factors (e.g. age at onset, presence/absence of family history, isolated vs. complex movement disorder) modulate the diagnostic yield and may be considered to guide selection of patients that are most likely to attain a genetic diagnosis using current genetic testing technologies [13, 14, 15]. For example, younger age at onset is generally associated with increased diagnostic yield in dystonia [15, 16], ataxia [17, 18], and PD [19, 20].

Regarding different sequencing technologies, WES and WGS both generally outperform targeted gene panels in terms of diagnostic yield due to broader coverage of the genome. Similarly, WGS is expected to return higher diagnostic yields than WES or custom gene panels due to coverage of intronic regions and improved ability to detect structural variants and repeat expansion disorders through various bioinformatic analyses [21, 22, 23, 24]. Although no studies have directly compared the diagnostic yield of WGS and WES in movement disorder cohorts, WGS has shown superior yield in other populations [25, 26].

Potential explanations for the large number of patients who remain genetically undiagnosed despite use of WES or WGS include: variants located in as yet undiscovered loci, complex variants that are difficult to resolve using current clinically validated sequencing technologies, variant mosaicism particularly where pathogenic variants are present in disease-related tissues but absent from blood, disorders due to polygenic or non-genetic causes and/or patients who have been phenotypically misclassified. For this population, periodic reanalysis of existing WES or WGS data may reveal new diagnoses that increase the diagnostic yield, without the need for repeat sequencing. These additional diagnoses may be achieved due to a combination of factors, including clarification and/or evolution of the patient phenotype, identification of new gene-disease associations, reclassification of variant pathogenicity and/or advancements in bioinformatic approaches. For example, reanalysis of WES data from a cerebellar ataxia cohort five years after initial analysis, resulted in variant reclassification in 24% of cases and a 7% increase in overall diagnostic yield, from 21% to 28% [27]. Similarly, application of a copy number variant analysis tool to existing WES data from a dystonia cohort increased the diagnostic yield by 1.5% [28].

Other Facets of Diagnostic Utility

Early use of genetic testing in the diagnostic workup of selected movement disorder patients may reduce time to diagnosis and provide cost savings compared to diagnostic strategies that do not emphasize early genetic testing. For example, targeted gene panel analysis in a cohort of dystonia patients was shown to be cheaper (€1822 per case vs. €2660 per case) and result in shorter time to diagnosis (28 vs. 102 days), when compared to theoretical diagnostic strategies (excluding use of NGS tests) formulated by movement disorder experts who were provided with detailed clinical features of the patients in this cohort [29]. Cost and resource savings from early genetic testing may in part derive from the avoidance of unnecessary and/or invasive investigations. For example, a genetic diagnosis of GCH1-related dopa-responsive dystonia could obviate the need for a lumbar puncture to assess cerebrospinal fluid neurotransmitter levels. Healthcare resource savings may also come from earlier diagnosis of, and thus earlier opportunities to treat, comorbidities that may not otherwise be considered in the absence of a specific genetic diagnosis (e.g., regular echocardiography to screen for development of cardiomyopathy in patients with Friedreich’s ataxia).

Targeted Medication

While medical management of movement disorders is often based on a syndromic diagnosis, a genetic diagnosis can facilitate a precision medicine approach, with use of specific medications or dosing regimens for individual genetic conditions. Recent examples of movement disorders with specific symptomatic treatments include the recently described spinocerebellar ataxia (SCA) 27B, where the use of 4-aminopyridine has been shown to reduce the severity and frequency of certain symptoms (e.g. diplopia, dizziness, and dysarthria) in up to 86% of patients [47, 48], and ADCY5-related dyskinesia, where the use of caffeine reduces paroxysmal hyperkinetic movements in up to 87% of patients, likely through the inhibition of adenosine A2A receptor activation of ADCY5 in the striatum [49]. In addition to symptomatic treatments, the number of disease-modifying therapies also continues to grow, e.g., omaveloxolone, an oral small molecule antioxidant that has recently been shown to slow progression and even improve neurological function in patients with Friedreich’s ataxia [50].

An increasing number of novel or repurposed therapeutic agents are in clinical trials for treatment of numerous different genetic movement disorders. Some examples of agents in phase III trials for movement disorders include BIIB122, an oral small molecule LRRK2 inhibitor in patients with LRRK2-related PD (NCT05418673) and metformin, which is being repurposed for trial use in patients with Huntington’s disease (NCT04826692).

Lastly, pharmacogenomic testing (genotyping of polymorphisms in genes that relate to drug response and metabolism) may be helpful in predicting the effect of medications in patients with movement disorders. As an example, testing for polymorphisms in genes encoding dopamine receptors and enzymes that metabolize levodopa could potentially be used to help predict efficacy of and likelihood of side effects to levodopa in individual patients with PD [51].

Deep Brain Stimulation

Although deep brain stimulation (DBS) has well-established efficacy in a variety of movement disorders, there is increasing evidence to show that response to DBS may vary depending on genetic factors. For example, although many monogenic forms of PD generally show good response to DBS [52], several case series of patients with GBA1 variants have shown a tendency towards more rapid cognitive impairment after DBS of the subthalamic nucleus (STN) [53, 54]. However, large prospective studies of DBS in patients with GBA1 variants stratified by variant type and DBS target (STN vs. globus pallidus interna [GPi]) are required to produce more definitive conclusions. Until such time, given that GBA1 patients experience favorable and durable motor response to DBS [55], genetic results should not be used as a determinant of DBS eligibility in monogenic PD. Instead, genetic status should be discussed with patients alongside other clinical factors that may impact response to and safety of DBS (e.g., age, comorbidities) to better inform discussions on likely outcomes.

Good motor outcomes from DBS of the GPi have been reported for multiple forms of monogenic dystonia including those related to variants in TOR1A, TAF1, SGCE, and KMT2B amongst others [56, 57, 58]. In contrast, DBS of GPi has been found to be generally ineffective in patients with ATP1A3-related dystonia-parkinsonism [59, 60, 61]. However, a few cases involving either a novel ATP1A3 variant or novel DBS target have shown reasonable response to DBS [62, 63] suggesting that, as is the case for PD, gene status alone should not be used to decide on eligibility for DBS.

Dietary Modification

Several genetic movement disorders respond to specific dietary modifications [44]. In particular, dietary vitamin supplementation may be indicated for genetic deficits in vitamin absorption or utilization e.g., vitamin E supplementation to stabilize and even improve symptoms in TTPA-related ataxia with vitamin E deficiency [64]. Conversely, dietary restrictions to reduce accumulation of toxic metabolites or to encourage use of alternative metabolic pathways may also be indicated for some disorders. An example of this is SLC2A1-related gluocose-transporter-1 deficiency where the ketogenic diet is used to generate ketone bodies that can provide the central nervous system with an alternative source of energy to glucose, thus preventing the seizures and paroxysmal dyskinesia that normally occur in this disorder [65, 66].

Gene Therapy

Gene therapy aims to alter gene expression or function for therapeutic purposes. Various techniques can be employed, including gene transfer to replace a dysfunctional gene, alteration of gene expression by up- or down-regulation, alteration of mRNA splicing or direct gene editing [67]. Currently, the only approved gene therapy for a genetic movement disorder is eladocagene exuparvovec, an intraputaminally-infused recombinant adeno-associated viral (AAV) vector containing the human DDC gene, for treatment of aromatic L-amino acid decarboxylase deficiency, a severe childhood-onset neurometabolic disorder that can present with oculogyric crisis and dystonia. Use of this gene therapy restores dopaminergic synthesis and results in rapid and robust improvements in cognitive and motor function [68].

Gene therapies for many other genetic movement disorders are either in development or clinical trials. Some examples include: LY3884961, an intra-cisterna magna injected AAV-based gene transfer therapy to restore glucocerebrosidase function in PD patients with pathogenic GBA1 variants (NCT04127578); BIIB094, an intrathecal antisense oligonucleotide designed to knockdown LRRK2 expression in PD patients with gain of function LRRK2 variants (NCT03976349); and PTC518, an oral small molecule splicing modifier that knocks down HTT expression in Huntington’s disease (NCT05358717). Successful development of gene therapies heralds exciting prospects for the treatment of genetic movement disorders.

Trigger Avoidance

For some movement disorders, a genetic diagnosis may provide guidance regarding medications and/or triggers that should be avoided. For example, although valproic acid is a commonly use medication to treat seizures and occasionally myoclonus, it may precipitate liver failure in and therefore is absolutely contraindicated for patients with POLG-related mitochondrial disease, which occasionally presents with various movement disorders including ataxia, parkinsonism, myoclonus, and dystonia [69]. Another example is the acute neurological deterioration seen in ATP1A3-related rapid onset dystonia-parkinsonism, which may be precipitated by environmental triggers, including alcohol, excessive exercise, and fever [44, 70]. Attempts to avoid relevant triggers should form part of management recommendations for patients with these genetic movement disorders.

Avoidance/Cessation of Unnecessary Treatment

Establishing a genetic diagnosis may facilitate discontinuation or avoidance of unnecessary treatments targeted at alternative diagnoses. For example, GCH1-related dopa-responsive dystonia typically presents with focal lower limb dystonia which, if not diagnosed, has been reported to occasionally be unnecessarily treated with orthopedic surgery aiming to correct the abnormal foot posture [71]. It is also not uncommon to encounter patients with a presumptive diagnosis of an autoimmune neurological disorder who are subsequently diagnosed with a genetic movement disorder, allowing for discontinuation of immunosuppressant medications [72, 73]. Although co-occurrence of a genetic movement disorder and an autoimmune neurologic disorder is well described [74, 75, 76], it remains unclear if these associations are causal (i.e., the genetic disorder predisposes to a secondary autoimmune process) or occur by chance.

Overall Therapeutic Impact

Despite being a critical component of the overall utility of genetic testing, the therapeutic impacts of establishing a genetic movement disorder diagnosis are reported in only a few diagnostic cohort studies. In a small cohort of pediatric-onset movement disorder patients, 80% (8/10) of genetic diagnoses had specific treatment implications [77] while in another pediatric movement disorder cohort, 38% (10/26) of individuals obtaining a genetic diagnosis received disease-specific treatment [78]. In a larger cohort of dystonia patients, 34.1% (46/135) of the genetic diagnoses were associated with disease-specific treatment (32/46) or screening (14/46) recommendations, although the number of patients in whom such recommendations were enacted and the outcomes of this were not reported [15]. Hence, there is a great need for longitudinal data incorporating objective disease outcome measures to quantify and qualify the long-term therapeutic impact of genetic testing in cohorts of movement disorder patients. This data would help to guide clinical decisions and guidelines surrounding genetic testing in movement disorder patients. Lastly, it should also be noted that the therapeutic impact of genetic testing is not static – it is likely to increase with time as more gene-specific treatments continue to emerge.

Technical & Data Interpretation Issues

Limitations in NGS methodology (e.g., difficulty characterizing GC-rich regions, highly repetitive or homologous regions and complex structural variants) contribute to suboptimal diagnostic yield. Fortunately, improvements in diagnostic yield are being gained through the use of newer genetic technologies (e.g., long-read sequencing, optical genome mapping) that address many of the technical limitations of NGS [1, 36, 94, 95]. Indeed, use of long-read sequencing has contributed to the discovery and/or validation of recently described novel repeat expansion loci as the cause of various movement disorders, including the GAA repeat expansion in FGF14 that causes SCA 27B [48, 96] and TTTCA/TTTTA repeat expansions that cause the familial cortical myoclonic tremor with epilepsy syndromes [97, 98]. It is likely there are more movement disorder-causing repeat expansion loci yet to be discovered.

Another issue relating to the use of high throughput genetic sequencing is the increasing identification of variants of uncertain significance (VUS), which are variants for which there is insufficient data to allow for classification as a pathogenic/likely pathogenic or benign/likely benign, and high impact variants in genes of uncertain significance (GUS), where there is no firmly established link between the identified gene and the phenotype of interest. In the era of high-throughput NGS, VUSs and GUSs are common (10.6% to 44.8% of patients in the studies listed in Table 1), occasionally even eclipsing the actual diagnostic rate. VUSs have the potential to be misinterpreted by patients and may elicit negative reactions (e.g., distress over uncertainty, disappointment, frustration) [99]. Reclassification of VUSs and GUSs may occasionally be possible based on further in silico analyses or familial segregation studies. However, in most cases, functional studies are required to assess the biologic effect of the specific VUS/GUS and provide sufficient additional evidence for or against pathogenicity. Currently functional studies are often low-throughput and time- and cost-intensive. Furthermore, selection of an appropriate functional assay and model system is often not straightforward [100]. Overall, solutions to this problem require greater global collaboration between clinical and research teams, improvements in bioinformatic analysis pipelines, generation and curation of more comprehensive variant databases, use of more diverse genomic references, and development of reliable, high-throughput functional assays [100, 101].

Logistic Issues

Issues relating to access, cost and funding/insurance coverage for genetic testing hinder the widespread and equitable use of genetic testing. Despite the falling costs of genetic testing with time, recent surveys of movement disorder physicians show that testing costs and limited access to genetic testing and genetic counselling remain major barriers to routine genetic testing of movement disorder patients around the world [102, 103, 104]. Solutions to this require building genetic and genomics workforce capacity at a local level and also collaborative efforts at a global scale (e.g., the Global Parkinson’s Genetics Program [105]).

With the growing volume of genetic data being generated, logistical issues relating to storage and analysis of large amounts of genomic data as well as data integrity and privacy concerns are becoming increasingly important. Both technologic advances and development of clearer public policy in this area are required to protect the privacy of genetic data moving forward [106].

Legal, Ethical and Psychological Issues

A potential barrier to more widespread use of genetic sequencing is the risk of genetic discrimination [107]. A survey of patients with PD identified that the impact of genetic testing on the ability to obtain health and life insurance was a common concern [108]. Across the world, different jurisdictions and health systems have enacted regulations at either a legislative or industry level in an attempt to mitigate discrimination by insurance providers based on genetic results [109, 110]. However, ongoing genetic discrimination despite these policies [111] indicate a need for continuing review and augmentation of these regulations. In a similar vein, thoughtful public policy is also required to regulate the emerging potential for use of stored genetic data for non-medical (e.g., forensic) purposes [112].

Some patients may experience stigma and/or psychological distress following receipt of a genetic diagnosis. In particular, a systematic review exploring the psychological impact of genetic diagnosis found that unlike most other groups of patients, patients with Huntington’s disease uniquely experienced negative psychological impacts (including depressive symptoms and suicidal ideation) following genetic testing [113]. However, it is possible these symptoms could, at least in part, reflect pre-existing affective disorders or psychiatric manifestations of the disease itself. Provision of adequate pre- and post-test education, genetic counselling and psychological support is required to help mitigate this risk [113].

Other practical issues that benefit from adequate genetic counselling are those relating to incidental and secondary genetic findings [114] and the potential for genetic testing to identify unexpected familial relationships (e.g., non-paternity, consanguinity). Clinical practice relating to management of incidental findings and return of secondary findings varies globally; adherence to clinical practice guidelines on this topic is important to minimize harm [115].