Genetic heterogeneity in amyotrophic lateral sclerosis and related disorders

Abstract

The overarching aim of this thesis is to clarify and further our understanding of the genetic causes of amyotrophic lateral sclerosis (ALS) and related diseases. It is hoped that achieving this can help bring clarity to patients, relatives and carers by improving genetic counselling and aiding in the design of clinical trials by improving patient stratification based on genetic background.

In the first research chapter of this thesis, a meta-analysis of all genetic variants previously reported in ALS and frontotemporal dementia (FTD) patients is performed. 3,114 variants in 356 genes were identified from a manual screen of the extant literature. Ultimately, 112 variants in 21 genes are found to cross the evidence threshold to be classified as pathogenic or likely pathogenic. This study also confirms the effect of reduced variant penetrance in ALS and FTD and finds that many variants exhibit significant geographic heterogeneity. A web application (alsftd.tcd.ie) is made available to provide all supporting evidence in an accessible format for clinicians, patients and researchers.

The second study in this thesis focuses on the identification of short tandem repeats (STRs) and repeat expansions (REs) in next-generation sequencing data. A benchmarking study of 7 tools is performed to assess their ability to correctly identify large REs, to accurately measure STRs and finally to compare results between whole-exome sequencing data and whole-genome sequencing data from the same patients. It is identified that many tools have good utility for identifying REs and accurately measuring STRs; however, no single tool provides perfect discrimination and the accuracy of results can be highly gene dependent. Consequently, it is advised that significant results observed from these tools should be subject to validation either with polymerase chain reaction or by taking a consensus approach with other tools. The lessons learned from the benchmarking study are applied to the study of 132 epilepsy patients, wherein no evidence is found supporting the pleiotropic role of REs known to cause other neurological diseases in the pathology of this disease.

Following the dual observations from the meta-analysis that the majority of ALS research has been performed in a small number of regions and that several genetic variants exhibit significant geographic heterogeneity, it is deemed beneficial to study the genetics of ALS in previously understudied populations. The third study in this thesis concerns the genetic screening of 126 Cuban ALS patients and 111 controls for pathogenic genetic variants. A low rate of the C9orf72 RE is observed. Interestingly the cohort does not carry SOD1, TARDBP or VAPB variants that are identified to be prevalent in North and South America.

The final research chapter examines the genetic basis of ALS and the related conditions FTD and primary lateral sclerosis (PLS) in Ireland. One PLS patient is found to harbour a previously unreported variant in the gene SPAST. Variants in the same amino acid have previously been reported to cause adult onset hereditary spastic paraplegia, a condition with significant clinical overlap with PLS. The genetics of ALS and FTD in Ireland are found to be distinct from the rest of the world by their absences. While rates of the C9orf72 RE are found to be similar other European countries, Irish patients lack genetic variants that are commonly observed elsewhere. Finally a study of related individuals, who are similarly affected with ALS or FTD, but who have discordant C9orf72 genotyping, is performed to further elucidate the basis of this discordance.