Introduction:
Hereditary spastic paraplegia (HSP) is characterized by lower extremity weakness and spasticity. HSP is genetically heterogenous, with over 80 types identified to date. Autosomal dominant HSP (AD-HSP) is thought to account for 75 to 80% of the cases, and variants in SPAST represent up to 40% of AD-HSP.  Despite increasing knowledge of the genetic causes of HSP, for many individuals with a clinical diagnosis of HSP, the genetic etiology remains undetermined.  Few studies have evaluated the yield of broad genetic testing in a large heterogenous cohort. We evaluated the yield of genetic testing in a cohort of individuals with suspected HSP who underwent sequencing and copy number variant (CNV) analysis.  

Methods:
We performed a retrospective review of de-identified data from consecutive individuals who underwent testing via a Spastic Paraplegia panel. Individuals were included if they were suspected of having HSP based on the clinical history provided. The panel included sequencing and CNV analysis by a targeted next-generation sequencing (NGS) assay. Target regions included coding exons (± 20bp from the intron/exon boundary) of up to 75 nuclear genes associated with spastic paraplegia, as well as up to 42 non-coding variants in these genes catalogued as disease-associated by HGMD and/or ClinVar. Gene content varied due to the addition of genes to the panel over time. Variant interpretation was performed in accordance with modified ACMG/AMP guidelines. An informative result in a gene was defined as the identification of a pathogenic or likely pathogenic variant(s) consistent with the reported phenotype and disease inheritance.  

Results:
A total of 533 individuals underwent testing; 70.5% (376/533) were adults (≥18 years of age) at the time of testing, and over half (56.8%, 303/533) were male. In all, 28.3% of individuals (151/533) received an informative test result. Most informative results (80.1%, 121/151) were in adults. Pathogenic and likely pathogenic variants were reported in 25 different genes. Autosomal dominant inheritance was observed in 67.5% (102/151) of individuals with informative results, and 68.6% (70/102) of those variants were in SPAST. Variants in CTNNB1, PAH, and C19orf12 were only seen in individuals tested at age 17 or younger. Overall, 11.3% (17/151) of informative results involved a CNV, ranging in size from a single exon to entire gene deletions. Seventeen percent (12/70) of the variants in SPAST were CNVs. Autosomal recessive inheritance accounted for 29.1% (44/151) and X-linked recessive inheritance accounted for 3.3% (5/151) of informative results. Of the autosomal recessive genes, SPG7 (n=18) and SPG11 (n=11) were most frequent. Two genes (ABCD1 and PLP1) accounted for all the X-linked cases. Non-coding variants contributed to <1% of informative results.  

Conclusion:
Approximately 28% of individuals who underwent testing for HSP using a broad gene panel received an informative result. The most frequent informative results were in SPAST, SPG7, and SPG11, which is consistent with the literature. The frequency of SPAST variants among those found to have AD-HSP was higher than previously reported. CNVs contributed to the overall diagnostic yield, particularly in the SPAST gene. The frequency of CNVs in SPAST was also higher than previously reported.