Genetic Evaluation of Hereditary Spastic Paraplegia

Assistant Professor of Neurology, Weill Cornell Medicine

SYNOPSIS: Hereditary spastic paraplegia (HSP) refers to a group of rare, clinically heterogenous degenerative neurogenetic disorders resulting in spasticity, gait impairment, and falls as the result of a length-dependent upper motor neuron degeneration. Next-generation sequencing with multigene panels or exome analysis can confirm molecular diagnosis of approximately 30% of HSP patients. Multigene panels can identify the common causative variants, variants on rarely involved genes, and structural rearrangements.

SOURCE: Méreaux J-L, Banneau G, Papin M, et al. Clinical and genetic spectra of 1550 index patients with hereditary spastic paraplegia. Brain 2022;145:1029-1037.

The diagnosis of hereditary spastic paraplegia (HSP) can be challenging because of wide clinical and genetic heterogeneity in addition to overlapping features common to other neurologic disorders. The prevalence of HSP is one to nine people per 100,000, and onset can range from infancy to old age. The hallmark of pure HSP is progressive lower extremity spasticity, gait impairment, and frequent falls. Clinical examination can reveal spastic tone in the legs, gait impairment, hyperreflexia, and a positive Babinski sign. Complex HSP can be associated with cognitive impairment, cerebellar ataxia, peripheral neuropathy, epilepsy, optic atrophy, dystonia, or parkinsonism. Symptom progression is variable, although most cases are slowly progressive with extended periods of relative stability.¹ HSP subtypes also can be designated by the mode of inheritance, with autosomal dominant (AD) transmission being the most common, followed by autosomal recessive (AR), mitochondrial, and, rarely, X-linked forms.

To date, more than 80 causative variants have been identified in various forms of HSP. Genetic analysis with next-generation sequencing, either exome sequencing or multigene panels targeting known genes, is essential to diagnosis. Both techniques are becoming more widely available for comparable cost. As opposed to targeted testing of few frequently associated HSP genes based on phenotype, multigene panels permit efficient identification of rare pathologic variants and structural rearrangements, which are common in HSP.

The authors reviewed 1,550 HSP index cases referred to the Neurogenetics Laboratory of the Pitie-Salpetriere University Hospital or recruited through the French SPATAX network between December 2013 and July 2019. Clinical and genetic data were extracted from an established national clinical and research database, with informed consent in accordance with national ethical rules and with local French research regulations.

The study evaluated a multigene panel that included 65 HSP-related genes and included an additional seven genes for the last 373 patients. Cases identified with the multigene panel as the first-line tool were compared to a cohort where the panel was employed as a second-line tool after targeted testing of the most frequent HSP genes. A panel was considered positive if a pathogenic or likely pathogenic variant was identified. Pathogenic variants, likely pathogenic variants, and variants of undetermined significance (VUS) were defined according to the American College of Medical Genetics classification. Additional analyses were performed when a single variant associated with AR HSP was identified. Whole exome sequencing (WES) was performed on 42 samples when there was no causative variant or copy number variant on the HSP panel.Statistical analysis was conducted using R studio 1.3.1073. Chi-square or Fisher’s exact test was used to compare qualitative variables. Welch test was used to compare continuous variables, such as median age of onset and disability score. P < 0.05 was considered significant.

Most of the 1,550 index patients occurred de novo (54%), followed by AD forms (21.9%), AR forms (9.9%), and X-linked inheritance (0.1%). Inheritance pattern was unknown or unavailable for 13.8%. Median age of onset was 25 years and median disease duration was 12.4 years. The pure HSP forms were less frequent than the complicated forms (43.9% vs. 56.1%, P < 0.001).

Gene panel testing was the first-line analysis in 809 patients compared to 741 patients who had the gene panel test as a second-line analysis following targeted testing of the most frequent HSP genes on the basis of clinical presentation. The first-line group had an older age of onset (32 years vs. 20 years, P < 0.001). The second-line group had a higher rate of consanguinity (14.8% vs. 4.5%, P < 0.001). Variants in the SPAST and SPG7 genes were most frequent and accounted for 9.2% and 4.8%, respectively. Variants in the KIF1A, ATL1, SPG11, KIF5A, and REEP1 genes were found in 147 cases (9.5%). The remaining 28 genes accounted for 0.06% to 0.97% of cases each. Expanded screening in an additional 373 patients identified a causative variant in one patient.

Identification of genetic disorders was similar when using the panel as first- or second-line analysis. The probability of detecting a causative variant was higher in patients with a clear inheritance pattern or familial recurrence: AD, 44%, and AR, 42.2% vs. undetermined, 26.6%, or de novo cases, 23.9%. Mutated genes were associated with AD inheritance in 58.9% and with AR in 39.2%. De novo mutations accounted for 1.7% of all mutated cases.

Overall, 661 causative variants were identified in 475/1550 (30.65%) subjects. Of these, 30 were structural variants and 631 were nonstructural variants (premature termination codon variants, 51.2%; missense variants, 48.8%). Not surprisingly, lower extremity spasticity and weakness was the most common clinical feature among all patients. There were no significant differences in clinical features among mutated and nonmutated cases, apart from sensory neuropathy being more frequent among mutated cases and upper limb spasticity being more common among nonmutated cases.

Among variant carriers, early-onset disease was associated more frequently with SACS, ATL1, and KIF1A variants, and late-onset was associated more commonly with SPAST and SPG7 variants. There was no difference in age of onset between the AD and AR forms. Patients with AR HSP had more severe symptoms and were more likely to have complicated disease. Cognitive impairment, upper limb spasticity, bulbar symptoms, cerebellar ataxia, ocular movement abnormalities, diminished visual acuity, extrapyramidal involvement, and abnormal brain magnetic resonance imaging (MRI) were encountered more frequently in patients with AR HSP. Epilepsy and peripheral neuropathy were similar among AR and AD forms.

Although no particular clinical sign was specific for a molecular diagnosis, certain constellations of signs were more often associated with a gene. The combination of epilepsy and neuropathy pointed to KIF5A mutations. Neuropathy and cognitive deficit guided toward KIF1A. Abnormal brain MRI was more frequent with SPG11, SACS, and SPG7 variants. WES analysis revealed causative variants in 11 of 42 patients with negative panels. The study demonstrated the efficiency of HSP-targeting sequencing panels as first-line genetic testing. Early onset, consanguinity, AD inheritance, or familial recurrence increases the power of next-generation sequencing in HSP diagnosis. Follow-up WES or whole-genome sequencing (WGS) for panel-negative cases can increase the diagnostic yield up to 50%. Additionally, WES or WGS may be appropriate first-line testing in patients with complex neurologic syndromes where HSP is only one of several clinical features.

Commentary

The diagnosis of HSP may be challenging because of the broad clinical and genetic heterogeneity and lack of defined genotype-phenotype associations. Strategic testing with first-line targeted sequencing panels can maximize the diagnostic yield by including rare variants and structural rearrangements. Reflexive WES or WGS analysis can identify additional variants in patients with negative panel tests.

A clear mode of inheritance and identification of early-onset disease may help increase the probability of detecting a causative variant. Certain constellations of neurological signs can correlate with specific HSP mutations. However, specific neurological signs cannot predict a specific mutation.

Reference

1. Boutry M, Morais S, Stevanin G. Update on the genetics of spastic paraplegias. Curr Neurol Neurosci Rep 2019;19:18.