Friedreich Ataxia (FRDA) via the FXN GAA Repeat Expansion

Friedreich ataxia (FRDA) is an autosomal recessive neurodegenerative condition caused by biallelic pathogenic variants in the FXN gene. The most common pathogenic variant is an expansion of a GAA trinucleotide repeat within intron 1. An estimated 96% of individuals with Friedreich ataxia have biallelic expansions (i.e., a pathogenic expansion is found in both copies of FXN). The remaining 4% of individuals are typically compound heterozygous for a pathogenic expansion on one allele and a different type of pathogenic variant on the other (Delatycki and Bidichandani. 2019. PubMed ID: 31494282).

FRDA is the most common inherited ataxia, accounting for 50% of all hereditary ataxia cases (Williams et al. 2023. PubMed ID: 33085346). An estimated 1 in 50,000 individuals in the United States and 1 in 40,000 worldwide have FRDA (Bidichandani et al. 1998. PubMed ID: 20301458; Williams et al. 2023. PubMed ID: 33085346). It is estimated that 1 in 100 Europeans is a carrier of an expanded FXN allele (Pandolfo et al. 2008. PubMed ID: 18852343).

Major clinical features of FRDA include progressive ataxia (ataxic gait and limbs), dysarthria, dysphagia, decreased proprioception, distal muscle weakness, peripheral sensory neuropathy, absent lower limb tendon reflexes, spasticity, scoliosis, pes cavus, and hypertrophic cardiomyopathy. For the vast majority of FRDA patients, gait ataxia (caused by spinocerebellar degeneration) is the first presenting clinical feature, which is later followed by slurred speech and upper-limb ataxia as spinocerebellar degeneration progresses (Bidichandani et al. 1998. PubMed ID: 20301458). Less common features found in some individuals with FRDA include diabetes mellitis, lower urinary tract symptoms (urinary frequency and urgency), nystagmus, and optic nerve atrophy.

Clinical features of FRDA typically present before the age of 25, with the mean age of onset between 10 and 15 years (Bidichandani et al. 1998. PubMed ID: 20301458; Cook and Guinti 2017. PubMed ID: 29053830). The age of onset, presence of leg muscle weakness/wasting, duration until wheelchair use, and prevalence of cardiomyopathy, pes cavus, and scoliosis are all inversely correlated with the length of the expanded GAA repeat. In patients with biallelic pathogenic expansions, the size of the smaller of the two expanded alleles has been shown to correlate with age of onset and clinical manifestations more strongly than the larger of the two alleles (Bidichandani et al. 1998. PubMed ID: 20301458; Filla et al. 1996. PubMed ID: 8751856; Schols et al. 1997. PubMed ID: 9448568; La Pean et al. 2008. PubMed ID: 18759347). Variable expressivity has been demonstrated in FRDA, as 12% of individuals with biallelic pathogenic expansions present with FRDA with retained reflexes (FARR). FARR is typically characterized by later age of onset and lower incidence of secondary skeletal involvement and cardiomyopathy (Bidichandani et al. 1998. PubMed ID: 20301458; Verma and Gupta. 2012. PubMed ID: 23242090). 

Normal FXN alleles contain 5 to 33 GAA repeats. Alleles that have 34 to 43 repeats are known as normal mutable or premutation alleles. Repeats of this size are not associated with a clinical phenotype; however, they have a higher propensity to expand in the subsequent generation. Borderline alleles have 44-65 repeats, which are also at risk for expansion in the next generation, and there has been inconclusive evidence about whether these individuals are clinically affected. Alleles with 66 or greater repeats are considered pathogenic. Individuals with two pathogenic expansions or compound heterozygous for a pathogenic expansion and another type of pathogenic variant are clinically affected with FRDA.

Given the autosomal recessive nature of FRDA, carriers of pathogenic alleles (one expanded and one normal allele) are clinically not affected but have a 50% risk of passing the pathogenic allele on to their offspring. Individuals who are carriers of premutation or borderline alleles are also at risk for passing on a pathogenic allele to their offspring due to the spontaneous expansion of repeats during meiosis; however the absolute risk of this occurring has not been well established (Schols et al. 1997. PubMed ID: 9448568; Montermini et al. 1997. PubMed ID: 9259271; Cossee et al. 1997. PubMed ID: 9207112). Of note, GAA repeat alleles including interruptions are much less likely to expand into the pathogenic range compared to GAA repeat alleles that are uninterrupted (Masnovo et al. 2022. PubMed ID: 35952488).

Testing for FXN repeat expansions is recommended for individuals suspected of having FRDA or those with a family history consistent with autosomal recessive inheritance. Advantages of testing include differential diagnosis from other ataxia-causing syndromes, molecular confirmation of diagnosis, and determining carrier status.

FRDA is an autosomal recessive neurodegenerative condition caused by biallelic pathogenic variants in the FXN gene, most commonly expansion of a GAA trinucleotide repeat within intron 1 of FXN. Sequence variants including missense, nonsense, and splice site variants have also been reported to be pathogenic (Alper and Narayanan. 2003. PubMed ID: 12878293). Loss of function is the main mechanism of disease; hypomorphic alleles arising from missense variants are typically associated with mild to moderate symptoms (Williams et al. 2023. PubMed ID: 33085346). Partial and full deletions of FXN have been reported (Aguilera et al. 2023. PubMed ID: 38041144; van den Ouweland et al. 2012. PubMed ID: 22691228; Sacca et al. 2013. PubMed ID: 23196337).

Overall, the presence of expanded GAA repeats in intron 1 of FXN results in transcriptional deficiency due to multiple mechanisms. For one, there is difficulty encountered in transcription by the RNA polymerase over larger repetitive sequences, thus resulting in reduced FXN mRNA production. Highly repetitive DNA sequences (sticky DNA) can frequently form secondary DNA structures that also block progression of the RNA polymerase. In addition, presence of an expanded GAA repeat has been shown to be associated with repressed chromatin epigenetic markers in the surrounding region, further contributing to the repression of FXN expression (Delatycki and Bidichandani 2019. PubMed ID: 31494282).

Penetrance is complete in individuals with two pathogenic expansions (greater than 66 repeats) and in compound heterozygotes with one pathogenic expansion in one allele and a pathogenic loss of function variant in the other allele (Bidichandani et al. 1998. PubMed ID: 20301458). Reduced penetrance may be observed in individuals with alleles in the borderline range (44-65 repeats). The smallest expansion size at which symptoms will present is not known (Bidichandani et al. 1998. PubMed ID: 20301458).

Most pathogenic sized repeats are inherited from a carrier parent, and de novo expansion is not a known phenomenon. However, offspring may inherit a pathogenic expansion from a parent who carries a normal mutable or borderline sized repeat due spontaneous expansion of repeats during meiosis (Schols et al. 1997. PubMed ID: 9448568; Montermini et al. 1997. PubMed ID: 9259271; Cossee et al. 1997. PubMed ID: 9207112).

FXN encodes the protein frataxin, which is a nuclear encoded mitochondrial protein involved in iron homeostasis and biogenesis of iron containing cofactors (iron sulfur clusters) in ATP production. Frataxin deficiency leads to accumulation of iron in the mitochondria and reduced iron sulfur cluster formation, resulting in defects within the oxidative transport chain and accumulation of free radicals, ultimately leading to cell death. The tissues most affected by frataxin deficiency are those that have low rates of cell division and depend primarily on aerobic metabolism—most notably cardiac muscle, pancreatic beta cells, and neural tissue (Alper and Narayanan. 2003. PubMed ID: 12878293). Mouse models of FRDA are imperfect due to embryonic lethality of biallelic loss of function variants in Fxn, the mouse homolog of FXN. Ectopic expression of human FXN containing pathogenic GAA expansions in Fxn-knockout mice demonstrated phenotypes like those of FRDA and notably had reduced frataxin mRNA and protein expression in the cerebellum, heart, and skeletal muscle compared to Fxn-knockout mice expressing wild-type FXN (Al-Mahdawi et al. 2006. PubMed ID: 16919418).

Currently, there is one FDA (Food and Drug Administration) approved drug, omaveloxolone, to treat FRDA in individuals aged 16 and older (Lee. 2023. PubMed ID: 37155124). Individuals with FRDA have a suppressed NRF2 pathway due to the reduced production of iron sulfur clusters, causing increased oxidative stress, mitochondrial dysfunction, and cell death (Abeti et al. 2018. PubMed ID: 30065630). Omaveloxolone is a small molecule drug that is used to activate the NRF2 antioxidant pathway to counter the effects of the increased oxidative damage observed in individuals with FRDA (Lynch et al. 2019. PubMed ID: 30656180; Lynch et al. 2021. PubMed ID: 33068037; Lynch et al. 2023. PubMed ID: 36444905).

This test for GAA repeat expansions in intron 1 of FXN has approximately 96% clinical sensitivity for molecular confirmation of FRDA. The remaining 4% of FRDA cases require NGS of FXN to identify sequence and copy number variants in addition to repeat expansion testing (Bidichandani et al. 1998. PubMed ID: 20301458).

This test is designed to only detect pathogenic expansions of a GAA trinucleotide repeat in intron 1 of the FXN gene and does not assess for repeat interruptions or other potentially pathogenic variants within FXN. Our assay involves two amplicon-length assays from both the 3’ and 5’ ends of the repeat region to determine the number of repeats, and a 5’ repeat-primed PCR assay with a locus specific primer. Pathogenic expansions (66 or greater repeats) are not able to be sized by amplicon-length analysis but are detected by the repeat primed assay. For this reason, sizing is not available for pathogenic expansions and are reported as “expanded” alleles.

Testing for sequence variants and copy number variants within the FXN gene can be ordered using our custom panel tool.