Introduction

Neurogenetic disorders (NGDs) include a vast group of diseases with marked phenotypic and genetic heterogeneity which can present in the pediatric and adult setting [1,2,3,4,5,6,7,8]. Adult-onset NGDs can be clinically grouped into several diagnostic subcategories, including movement disorders, neuromuscular disorders, neuropathies, neurodegenerative disorders, mitochondrial disorders, metabolic disorders, neurotransmitter disorders, and epilepsies. Affected patients and their families often go through a lengthy diagnostic process as establishing a definitive diagnosis requires extensive clinical, radiological, and genetic investigations. Clinically, factors such as reduced penetrance and variable expressivity can make it challenging to identify a NGD based solely on family history [9].

Next generation sequencing (NGS) and large-scale genomic analyses have revolutionized diagnostics and led to advances in understanding the underlying genomic contribution to disease, particularly in the realm of NGDs [1, 3, 5, 8, 10]. These diagnostic tools have created opportunities to integrate genomic data into clinical practice, improve the diagnostic process and solve the problem of the lengthy diagnostic odysseys in rare NGDs [1, 5, 10, 11]. Precise and timely diagnosis of these disorders is essential to offer appropriate treatment and accurate genetic counseling.

The utility and diagnostic yield (DY) of NGS in monogenic NGDs have been described in numerous studies [1, 5, 10]. Global consensus suggests that targeted NGS gene panels can be used as a first-tier test for patients with a suspected complex NGD, and if results are negative, exome sequencing or genome sequencing can be considered as second tier testing platforms [5, 11]. In a study by Marques Matos et al. [12], ~33.2% of patients, including adults and children with neurological disorders of undetermined etiology, received a definitive diagnosis by using NGS panels [12].

African populations harbor the greatest human genomic diversity, but despite this, genomic studies, and current knowledge on NGDs affecting these populations are limited [3]. In low-middle income countries (LMICs) the availability of NGS is largely limited by factors such as cost and lack of infrastructure and expertise in the public health care system. The South African public health sector provides medical care to the majority of South Africans, and the limited availability of comprehensive genetic testing decreases the overall DY in adults affected by NGDs.

The multidisciplinary adult neurogenetic clinic at Tygerberg Hospital (TBH) in South Africa has had access to NGS panel diagnostic testing since 2019. Here, we present the results of an audit of this service over a 4-year period and highlight DYs and recommendations for incorporating NGS testing into other LMIC neurogenetic services.

Materials and methods

Study design and setting

This was a retrospective observational descriptive study, conducted in the adult neurogenetic clinic of TBH, Cape Town, South Africa. This academic hospital is the largest state hospital in the Western Cape and the second largest medical facility in the public healthcare sector in South Africa, providing medical care to a large and diverse low- to middle income population. Adult patients with suspected NGDs, are referred to the TBH adult neurology service from primary, secondary, and tertiary level care, as well as from the private sector. If an NGD is suspected based on family history, medical history, clinical signs, symptoms, and neuroimaging findings, the patient is evaluated in the neurogenetic clinic by a multidisciplinary team consisting of Neurologists, Medical Geneticists and Genetic Counselors. NGS panel testing is requested in patients where the differential diagnosis points to a NGD amenable to diagnosis by NGS panel. When applicable, a clinical assessment for at risk family members is arranged to determine whether they are affected and offer diagnostic, predictive, carrier or VUS resolution testing.

Study sample

The study included all adults (18 years and older) who underwent NGS panel testing during the 4-year study period (from January 1, 2019, to December 31, 2022) through the adult neurogenetic clinic at TBH. During this period, the adult neurology outpatient service assessed an estimated total of 4800 new patients, of whom 240 patients with suspected NGDs (60 per year) were evaluated by a multidisciplinary team at the neurogenetic clinic.

Data collection and analysis

Following pre-test genetic counseling and informed consent, genomic DNA was collected using buccal swabs, assisted saliva kits, or venous blood draws. NGS panel testing was performed by Invitae Corporation (San Francisco, California, USA), a CAP-accredited and CLIA-certified clinical diagnostic laboratory. The analysis included full-gene sequencing and deletion/duplication (CNV) analysis using NGS technology aligned to the GRCh37 reference genome. The assay has demonstrated >99% analytical sensitivity and specificity for single nucleotide variants, small insertions/deletions (<15 bp), and exon-level CNVs based on validation studies.

Participants were selected for testing following multidisciplinary evaluation, based on clinical suspicion of a NGD. Test results were categorized as: (i) positive-affected, when a pathogenic or likely pathogenic variant consistent with the phenotype was identified; (ii) positive-carrier, when a single pathogenic or likely pathogenic variant was found in a gene for an autosomal recessive condition without a second causative variant; (iii) inconclusive, when only a VUS was detected; and (iv) negative, when no clinically significant variants were identified.

Certain neurogenetic conditions that require alternative diagnostic methods such as repeat expansion disorders (e.g., Huntington’s disease, myotonic dystrophy, spinocerebellar ataxias) and PMP22 duplications associated with Charcot-Marie-Tooth disease type 1 A, were evaluated separately through the South African National Health Laboratory Service (NHLS). These local diagnostic results were not included in the current study. Details of NHLS available tests and turnaround times are provided in Supplementary Table 1.

Results

Participant characteristics

This study included 74 participants: 43 clinically affected index cases who underwent NGS panel testing, and 31 family members who received targeted testing for known familial variants. Genetic testing was pursued based on a range of neurological indications, which were grouped into major clinical categories to guide panel selection. These categories included neuromuscular disorders (e.g., muscular dystrophies, hereditary neuropathies), movement disorders (e.g., hereditary ataxias, dystonias, early-onset parkinsonism), and adult-onset neurodegenerative conditions (e.g., leukodystrophies, hereditary spastic paraplegias, frontotemporal dementia). Testing was primarily performed using a range of Invitae diagnostic panels tailored to these phenotypes, such as the Comprehensive Neuropathies panel, Neuromuscular Disorders panel, Dystonia Comprehensive panel, and the Hereditary Amyotrophic Lateral Sclerosis, Frontotemporal Dementia, and Alzheimer Disease panel. A detailed list of panel types, associated disorder groups, and included genes is provided in Supplementary Table 2.

Table 1 presents the demographic characteristics of the study cohort. At the time of NGS panel analysis, the median age of participants was 36 years (interquartile range: 25–53 years), with ages ranging from 18 to 75 years. The cohort exhibited a female-to-male ratio of approximately 1.5:1. Self-reported ancestry indicated that the majority of participants identified as Caucasian (59.5%), followed by individuals of Mixed Ancestry South African (24.3%), African (9.5%), and Indian (6.7%) descent.

Table 1 Participant characteristics.
Full size table

Diagnostic yield in index cases

The diagnostic yield of NGS panels in this cohort is summarized in Tables 1 and 2. The overall diagnostic yield across the 43 index cases was 39.5% (17/43). Among the 22 Neuromuscular disorder panels requested, 10 returned a definitive molecular diagnosis, yielding a diagnostic rate of 45.5%. Similarly, 4 out of 19 neuropathy panels resulted in a confirmed diagnosis, corresponding to a DY of 21%. No pathogenic or likely pathogenic variants were identified in panels requested for movement disorders, developmental brain abnormalities, or metabolic disorders. This is consistent with the lower pre-test probability for these conditions, as they are less commonly monogenic in origin compared to neuromuscular disorders. However, the interpretation of DYs in specific phenotypic subgroups is limited by the small sample sizes. Carrier status for pathogenic variants was detected in 7 of the 43 cases (16.3%), while VUSs were the only findings in 12 cases (27.9%). In the remaining 7 cases (16.3%), NGS panel analysis returned negative results.

Table 2 Phenotype and pathogenic variants in affected index cases.
Full size table

Pathogenic variants identified in index cases

Table 2 summarizes the clinical features and pathogenic variants identified in participants for whom a definitive molecular diagnosis was established. Of the 17 individuals with confirmed diagnoses, 52.9% (9/17) had conditions inherited in an autosomal dominant manner, while 41.2% (7/17) were affected by autosomal recessive disorders, comprising 3 cases with homozygous variants and 4 with compound heterozygous variants. One participant (5.9%) was diagnosed with an X-linked condition.

Specific case examples

To illustrate the clinical and diagnostic value of genomic testing, several representative cases from the cohort are described in more detail below:

Participant 1 (Table 2) is a female of European ancestry, who presented with asymmetric scapular winging and muscle weakness affecting her face, neck flexors, and distal limbs. She exhibited subtle glove-stocking sensory changes, although her nerve conduction studies were normal. A novel heterozygous pathogenic deletion (Exons 14–77) in NEB was identified through the incorporation of CNV analysis tools into NGS variant-calling pipelines. Her family history strongly suggested an autosomal dominant inheritance pattern, as her mother, brother, and maternal grandmother were similarly affected. The same pathogenic variant was confirmed in her mother and brother. However, at the time of testing, only preliminary evidence supported the link between this variant and autosomal dominant nemaline myopathy, leading to her initial classification as a carrier for autosomal recessive NEB-related conditions. Since then, evidence has emerged showing that structural variants in NEB can cause an autosomal dominant condition, likely through a dominant-negative mechanism, with numerous families identified globally [13].This case underscores the evolving nature of variant classification, highlighting how variants can be reclassified as benign or pathogenic as more evidence becomes available. It also emphasizes the critical role of integrating CNV analysis tools into NGS pipelines, particularly in LMICs where alternate CNV techniques may not be available. This approach not only enhances DY without the need for additional technologies like chromosomal microarrays but also enables the identification of novel variants in underrepresented populations.

The clinical presentation of Participant 2 (Table 2) was consistent with suspected retinal vasculopathy and cerebral leukoencephalopathy with systemic manifestations (RVCL-S). The TREX1, Exon 2, c.795_802dup (p. Arg268Lysfs*12) variant was initially classified as a VUS. However, given that previously reported variants associated with this condition are also heterozygous frameshift variants and based on ACMG criteria (PSV1 and PM2), the variant is classified as likely pathogenic and is considered causative for her phenotype [14].

Participant 3 (Table 2) presented with early-onset, slowly progressive muscle weakness. The Invitae Limb-Girdle Muscular Dystrophy panel identified a heterozygous likely pathogenic variant in TTN. This sequence change creates a premature translational stop signal, and it is expected to create a truncated TTN protein. TTN is the largest gene in the human genome and is associated with a spectrum of autosomal dominant and recessive cardiac and neuromuscular conditions. However, this case is not considered solved by the detected TTN variant, as her phenotype does not align with tardive tibial muscular dystrophy, which is characterized solely by a distal lower limb phenotype. Autosomal dominant hereditary myopathy with early respiratory failure has only been associated with gain of function TTN pathogenic variants located in the 119th fibronectin-3 domain. This case underscores the critical importance of genotype-phenotype correlation. Even when a report identifies a pathogenic variant or suggests a diagnosis, the variant may not explain the clinical presentation completely, or in part and could be incidental. Additionally, these cases highlight the importance of involving a multidisciplinary team including neurologists, medical geneticists, and genetic counselors, in interpreting results to ensure accurate diagnoses and appropriate genetic counseling.

Importantly, molecular genetic diagnoses create opportunities for precision treatment.

A timely diagnosis of Pompe disease in Participant 4 (Table 2) was pivotal, enabling the initiation of enzyme replacement therapy, which has the potential to significantly modify the disease course and enhance the patient’s quality of life. Notably, this diagnosis represents the only case in this study where a disease-modifying treatment was explicitly implemented. However, a critical point for discussion is that precision treatments are on the horizon, albeit potentially distant in South Africa, and obtaining a precise genetic diagnosis positions clinicians to assemble trial-ready cohorts for future therapeutic advancements.

Targeted family testing results

Among the 31 participants who underwent targeted familial variant NGS testing, 4 were clinically affected, all of whom received a confirmed molecular diagnosis through diagnostic testing. Predictive testing was carried out for 11 asymptomatic individuals, resulting in 2 confirmed diagnoses. Carrier status screening was conducted for 13 individuals, while testing to resolve VUSs was performed for 3 participants. Table 1 provides a summary of the indications for familial variant testing among these 31 relatives and their relationships to the index cases.

Discussion

This study aimed to evaluate the DY of NGS gene panels in a South African adult patient cohort with suspected NGDs attending a tertiary neurogenetics service in an LMIC setting. The overall study DY was 39.5% (17/43 patients). This is comparable with the mean diagnostic yield of 25.6% detected in other studies conducted in high-income countries (HICs). Several studies have investigated the diagnostic yield of molecular testing in adult neurogenetic disorders. In a study by Marques Matos et al. [12], approximately 33.2% of patients, including adults and children with neurological disorders of undetermined aetiology, received a definitive diagnosis by using NGS panels [12]. Winder et al. [15], conducted a study to evaluate the clinical utility of NGS multigene analysis in over 25,000 patients (<1 – 96 years) with neuromuscular disorders. By using NGS-based gene panels, a definitive molecular diagnosis was obtained for 5055 of the 25,356 individuals. This represents an overall diagnostic yield of 20%, and a range of 4–33%, depending on the panel used in individuals with neuromuscular diseases [15]. Kwan et al. [16] did a mini-review study to describe the use of gene panels in the diagnosis of neuromuscular disorders in children and adults. They found that in the studies using comprehensive gene panels for undifferentiated categories of neuromuscular disorders, the yield has ranged from 12.9% to 48.8% [16]. A systematic review done by Gorcenco et al. 2020, identified 28 studies that evaluated the efficacy of NGS in diagnosing movement disorders in children and adults [17, 18].

Limited studies in other LMICs are available for NGDs in adult cohorts. Ganapathy et al. [19] reported an overall diagnostic yield of 40%, however most of the cohort were pediatric patients. Data from their study strongly suggested that the NGS-based multi-gene tests should be considered as the first-tier genetic test for neurological disorders in India [19] (Table 3).

Table 3 Diagnostic yield for NGS panels in neurogenetic disorders in other countries.
Full size table

In this study, 16.3% (7/43) of index cases were found to carry heterozygous pathogenic variants in genes associated with autosomal recessive conditions. These findings are most consistent with carrier status, rather than a definitive molecular diagnosis. However, such results may still hold clinical relevance, particularly when the individual’s phenotype partially aligns with the associated condition. In these cases, a second pathogenic variant may have been missed due to test limitations (e.g., deep intronic or regulatory variants not covered by the panel), or may have been detected but classified as a VUS.

Participant 18 (Supplementary Table 3) exemplifies this scenario, where a single VUS may still carry potential diagnostic importance depending on future reclassification or emerging evidence. She presented with ataxia, spasticity, and cognitive decline beginning at age 23, along with diffuse cerebral white matter loss on MRI, features suggestive of a leukodystrophy. Molecular testing identified two variants in the EIF2B5 gene: one likely pathogenic splice site variant (c.1654+1G>T (splice donor) and one missense VUS (c.383A>G (p.Tyr128Cys)). While a definitive molecular diagnosis could not be made, the clinical and genetic findings are suggestive of EIF2B5-related vanishing white matter disease. This case highlights how VUS findings, particularly when occurring alongside a known pathogenic variant, may hold evolving diagnostic significance, warranting close follow-up and potential reclassification over time.

In this study, a total of 67.4% (29/43) index cases had at least one VUS detected. In 27.9% (12/43) of index cases, only VUSs were reported, and a definitive diagnosis could not be achieved. This may be explained by our population being underrepresented in global databases and emphasizes the importance of further studies, and creation of local genomic databases. Participant 19 (Supplementary Table 3) is an African male who presented with slowly progressive spastic paraparesis from the age of 13 years, and he was homozygous for the c.1507A>G (p. Lys503Glu) variant in CYP7B1, which was classified as a VUS. His phenotype was compatible with autosomal recessive hereditary spastic paraplegia type 5 A, associated with bi-allelic variants in the CYP7B1 gene. His asymptomatic mother was found to be a heterozygous carrier for the same variant, but his father was not available for testing. Family testing to try and resolve variants of uncertain significance was limited due to unavailability of parents/other appropriate family members.

Conclusion

To the best of our knowledge, this is the first study that specifically investigated the DY of commercially available NGS panels in a cohort attending a tertiary adult neurogenetic service in an LMIC. The high overall DY in this study is comparable to previously reported adult neurogenetic cohorts in HICs. This study illustrates that the use of NGS panel testing can be successfully implemented in the clinical practice of LMICs and assist in the diagnostic workup in patients affected by complex NGDs. The high detection rate of VUSs highlights that our population is underrepresented in global databases, and further studies are needed to resolve these variants. A definitive diagnosis is the cornerstone of clinical management, and we are hopeful that NGS panel testing will be incorporated as a useful diagnostic modality in neurology practice in South Africa.