Long-read sequencing approaches to unravel the complex architecture of RFC1 repeat expansion associated with Cerebellar Ataxia, Neuropathy, and Vestibular Areflexia Syndrome (CANVAS)

Cerebellar Ataxia, Neuropathy, and Vestibular Areflexia Syndrome (CANVAS) is a late-onset, autosomal recessive neurodegenerative disease, clinically manifested as a combination of cerebellar ataxia, sensory neuronopathy, and bilateral vestibular areflexia. In 2019, the aetiology of CANVAS was attributed to a biallelic intronic repeat expansion (RE) of the AAGGG pentanucleotide in the RFC1 gene, differing in size and nucleotides combination from the wild-type allele (AAAAG)11 reported in the reference human genome. Thereafter, the genotypic spectrum has been broadened to include both pathogenic and non-pathogenic alternative repeat motifs, motif-specific pathogenic thresholds of expansion, and compound heterozygous configurations with point mutation alleles. Moreover, many uncertain significance alleles have been reported to date, whose pathogenic role is currently unknown. The diagnosis of CANVAS relies on routine laboratory techniques such as Flanking PCR and Repeat-Primed PCR. However, these standard methods have significant limitations in terms of RE sizing and motif composition determination. Consequently, they often result in incomplete or inconclusive outcomes, so a growing number of diagnostically challenging cases have been identified over time. The present thesis project focuses on a subset of 40 cases (34 affected patients and 6 unaffected relatives) with inconclusive diagnostic test results. The aim is to overcome current limitations by optimizing the diagnostic workflow through the integration of conventional PCR-based methods with Long-Read sequencing approaches like Nanopore sequencing (Oxford Nanopore Technologies, ONT) and Single-Molecule Real-Time sequencing (Pacific Biosciences, PacBio). By applying these strategies, 38/40 previously unsolved cases have been carefully analysed through Long-Read sequencing approaches to define patient-specific RFC1 RE architecture. For 4/40 patients (all affected), complete analysis was not possible due to poor DNA quality/quantity. The optimized workflow allowed to identify biallelic pathogenic RFC1 RE in 26.7% (8/30) of the affected patients and to define carrier status in 30.6% (11/36, including 7/30 affected patients and 4/6 unaffected relatives) of analysed individuals. Moreover, in the 41.7% (15/36, including 12/30 affected patients and 3/6 unaffected relatives) of subjects, the refined workflow enabled the identification of novel allelic configurations harbouring complex motif patterns, which are still associated with uncertain pathogenic significance. In conclusion, the integrated workflow proposed in this thesis improves the diagnostic rate in RFC1/CANVAS analysis and provides new insights into genotype-phenotype correlations, contributing to a deeper comprehension of the molecular architecture underlying RFC1 RE in CANVAS disease.