Clinical Description
To date, more than 400 individuals with GAA-FGF14-related ataxia have been identified [Ashton et al 2023, Bonnet et al 2023, Brais et al 2023, Hengel et al 2023, Iruzubieta et al 2023, Novis et al 2023, Pellerin et al 2023a, Rafehi et al 2023, Wilke et al 2023, Wirth et al 2023, Zeng et al 2023, Ando et al 2024, Borsche et al 2024, Méreaux et al 2024, Pellerin et al 2024b, Pellerin et al 2024c]. The following description of the phenotypic features associated with this condition is based on these reports (see Table 2).
GAA-FGF14-related ataxia is a mid to late adult-onset slowly progressive cerebellar syndrome with predominant gait involvement. Age of onset and clinical presentation can vary within the same family.
The median age at onset is 60 years (range: 21 to 87 years) [Iruzubieta et al 2023, Pellerin et al 2023a, Rafehi et al 2023, Wilke et al 2023, Wirth et al 2023, Zeng et al 2023, Méreaux et al 2024]. While the most common manifestation at disease onset is an unsteady gait (80%), nearly 50% of individuals present initially with episodic manifestations, including ataxia, visual disturbances (diplopia, oscillopsia, visual blurring), vertigo, or dysarthria, on average two to four years before the onset of progressive ataxia [Ashton et al 2023, Bonnet et al 2023, Pellerin et al 2023a, Pellerin et al 2024c]. The frequency and duration of these episodes are highly variable: they may last from minutes to days and occur daily to monthly [Ashton et al 2023]. Alcohol intake and physical activity are common triggers [Bonnet et al 2023, Pellerin et al 2024c]. Caffeine has also been reported to trigger episodes [Ashton et al 2023].
Although some individuals eventually require assistance with mobility, the use of a wheelchair is uncommon even after protracted disease duration.
Dysarthria, which does not develop in all individuals (40%-60%), often remains mild to moderate [Iruzubieta et al 2023, Pellerin et al 2023a, Wilke et al 2023, Méreaux et al 2024, Pellerin et al 2024c]. Speech remains easy to understand in most individuals, although occasional words may be unintelligible.
Dysphagia develops in less than half of individuals. Although it may increase the risk of aspiration, dysphagia is very rarely severe enough to require enteral nutrition or cause cachexia.
Visual disturbances, such as diplopia, oscillopsia, or visual blurring, are common.
Cerebellar oculomotor signs, which may include horizontal gaze-evoked nystagmus, downbeat nystagmus, dysmetric saccades, saccadic pursuit, rebound nystagmus, and impaired visual fixation suppression of the vestibuloocular reflex, occur in almost all individuals. Horizontal gaze-evoked nystagmus and downbeat nystagmus are common. Downbeat nystagmus may present episodically or occur at disease onset with other cerebellar oculomotor signs in the absence of other neurologic findings [Pellerin et al 2024b].
Vertigo and/or dizziness may occur during episodes of ataxia or interictally. It may lead to gait unsteadiness.
Unilateral or bilateral vestibular hypofunction is common [Pellerin et al 2023a, Rafehi et al 2023, Wilke et al 2023, Pellerin et al 2024b, Pellerin et al 2024c]. Some individuals experience gait unsteadiness and dizziness due to vestibular hypofunction.
Postural or rest tremor of the upper limbs is observed in some patients [Pellerin et al 2023a, Wilke et al 2023, Wirth et al 2023, Méreaux et al 2024].
Afferent sensory deficit manifests as reduced vibration sense and hyporeflexia; however, sensory neuropathy is not commonly demonstrated on nerve conduction studies [Iruzubieta et al 2023, Pellerin et al 2023a, Rafehi et al 2023, Wilke et al 2023, Pellerin et al 2024c]. When peripheral neuropathy is present, nerve conduction studies are consistent with mild axonal sensory or sensorimotor polyneuropathy [Pellerin et al 2023a, Wirth et al 2023, Pellerin et al 2024c]. Whether the polyneuropathy is pathophysiologically related to GAA-FGF14-related ataxia or simply reflective of an age-related disease process remains to be established. Muscle stretch reflexes can be normal, decreased, or brisk.
Spasticity of the lower limbs is most often mild and is not a common manifestation.
Parkinsonism is uncommon (4%-12%) [Rafehi et al 2023, Wilke et al 2023, Pellerin et al 2024c].
Autonomic dysfunction is rare at disease onset but may develop later in the disease course. Urinary urgency and erectile dysfunction occur in 28%-57% of individuals and 13% of males, respectively. In comparison, population-based surveys in the United States have shown that the prevalence of urinary urge incontinence ranges from 1.7% to 36.4% of the general population [Milsom et al 2014], and cross-sectional studies have revealed a prevalence of erectile dysfunction of 18.4% of adult males ages 20 years or older [Selvin et al 2007].
Frank autonomic dysfunction manifested by orthostatic hypotension is rare in GAA-FGF14-related ataxia [Wilke et al 2023, Wirth et al 2023, Méreaux et al 2024, Pellerin et al 2024c].
Hearing loss, namely presbycusis, has been described in some individuals [Rafehi et al 2023].
Cognitive impairment is relatively infrequent, even in advanced stages of GAA-FGF14-related ataxia [Wilke et al 2023].
Of note, age-related mechanisms as well as other additional diseases (age related or not age related; acquired or inherited) can contribute to or aggravate the clinical features of GAA-FGF14-related ataxia [Wilke et al 2023, Pellerin et al 2024b].
The possibility of concurrent medical illnesses, which are common in the elderly, must be considered in late-onset diseases such as GAA-FGF14-related ataxia. It has been shown that these are frequent and add to the neurologic spectrum and disease burden of underlying hereditary late-onset ataxia [Wilke et al 2023, Pellerin et al 2024b].
Life span does not appear to be shortened in individuals with GAA-FGF14-related ataxia [Wirth et al 2023].
Intergenerational Instability
The FGF14 GAA repeat is highly unstable and almost always changes in size upon parent-to-offspring transmission when the size of parent's GAA repeat expansion is greater than 75 repeats [Pellerin et al 2024b].
The size of the GAA repeat is more likely to expand with maternal transmission and to contract with paternal transmission [Pellerin et al 2023a, Pellerin et al 2024b].
The instability of the GAA repeat locus upon maternal transmission, which is at high risk of further expansion, partly accounts for the high incidence of simplex cases of GAA-FGF14-related ataxia (i.e., a single occurrence of a disorder in a family), whereby an unaffected mother transmits an expanded pathogenic allele to her offspring.
In contrast, contraction of the size of the GAA repeat upon male transmission may lead to transmission of reduced-penetrance alleles to the offspring, resulting in "generation skipping" of the disease [Pellerin et al 2023a]. This differential transmission dynamic also likely accounts for the reduced male transmission of the disease observed in two studies [Pellerin et al 2023a, Méreaux et al 2024].
The degree of intergenerational instability is proportional to the size of the GAA repeat of the transmitted allele and dependent on the purity of the repeat tract. GAA repeat expansions may be pure (GAA)n in sequence or may be interrupted with regions of non-GAA sequences. During intergenerational transmission, pure GAA repeats have been shown to be more unstable than non-GAA-pure repeats. Only pure GAA repeats are believed to be pathogenic, while non-GAA-pure repeats are believed to be not pathogenic for ataxia [Hengel et al 2023, Pellerin et al 2023b].
Prevalence
The prevalence of GAA-FGF14-related ataxia is difficult to estimate given that only about 400 individuals have been reported to date. The prevalence of late-onset cerebellar ataxia of unknown cause is one to nine in 100,000 individuals (see Orphanet). GAA-FGF14-related ataxia has been identified in cohorts with adult-onset ataxia of previously unknown cause at rates ranging from 9% to 61%.
Excluding French Canadian cohorts, we have estimated the prevalence of GAA-FGF14-related ataxia to be in the range of 0.1 to three in 100,000 individuals of European ancestry [Pellerin, Danzi, Renaud, Houlden, Synofzik, Zuchner, & Brais, personal observation]. This prevalence, like that of spinocerebellar ataxia 1, 2, 3, and 6, has recently been validated by a systematic comparative study of consecutive patient cohorts in a single-center European study [Hengel et al 2023]. Although a large degree of uncertainty remains regarding the prevalence of GAA-FGF14-related ataxia, to date it appears to be among the most common causes of inherited adult-onset ataxia as well as of autosomal dominant ataxia (of any age) [Hengel et al 2023]. However, GAA-FGF14-related ataxia may not be as common in East Asian populations, as it was not identified in 312 patients with suspected spinocerebellar degeneration of unknown cause from Hokkaido Island in northern Japan in one study [Mizushima et al 2024], and was identified in only 11 of 940 individuals (1.2%) from Japan with chronic progressive cerebellar ataxia in another study [Ando et al 2024].
It is important to note that although the frequency of FGF14 alleles longer than 250 triplets is estimated to be about 1%-2% in the population [Pellerin et al 2023a, Rafehi et al 2023, Méreaux et al 2024, Pellerin et al 2024a], this does not reflect a prevalence of GAA-FGF14-related ataxia nearing 1-2 in 100 individuals. In fact, 88.9% of alleles longer than 250 triplets in 2,191 controls undergoing long-read sequencing were identified to be non-GAA-pure and seem to have no association with GAA-FGF14-related ataxia [Pellerin et al 2024a].
Geographic variation in prevalence. Rates of GAA-FGF14-related ataxia are reported to be as high as 60% in individuals of French Canadian descent with adult-onset ataxia of previously unknown cause [Pellerin et al 2023a], making GAA-FGF14-related ataxia one of the most common genetic causes of late-onset ataxia within the French Canadian population [Brais, personal observation]. The enrichment in this population is likely due to a founder effect, as such effects have been previously described in this population [Scriver 2001], and some French Canadian individuals with GAA-FGF14-related ataxia appear to share a common haplotype at the FGF14 locus [Pellerin et al 2023a].