Frequency of GAA-FGF14 Ataxia in a Large Cohort of Brazilian Patients With Unsolved Adult-Onset Cerebellar Ataxia

Results

We identified 8 index patients (8/93; 9%) who carried an FGF14 (GAA)_≥250 expansion (Figure). The expansion was also identified in the affected son (patient 1.2) of an affected mother (patient 1.1). In this family, the transmission of an expanded allele resulted in expansion in the female germline (expansion from 253 to 268 repeats), consistent with previous reports.^1,2,7 We also identified a patient carrying biallelic GAA repeat expansions (363 and 448 repeats). Ancestry analysis of short-read sequencing data⁸ showed that 1 patient was of admixed American ancestry and 2 patients were of European ancestry. One patient self-reported being of African descent, while the remaining patients self-reported being of European descent.

• Download figure
• Open in new tab
• Download powerpoint

Figure Allele Distribution and Genetic Analysis of the FGF14 GAA Repeat Locus

(A) Allele distribution of the FGF14 repeat locus in 94 Brazilian patients with unsolved adult-onset ataxia. The repeat length was measured by capillary electrophoresis of fluorescent long-range PCR amplification products. The allele sizes are expressed in numbers of triplet repeat units. Expanded alleles consisting of non-GAA repeats are represented by red triangles. The dashed red line indicates the pathogenic threshold of at least 250 GAA repeat units. The bold dashed black line within the violin plot shows the median allele size. (B) Fragment length analysis results of patient 1.1 carrying an expanded allele of 253 GAA repeat units. (C and D) Results of repeat-primed PCR targeting the (C) 3′-end and (D) 5′-end of the FGF14 repeat locus in patient 1.1. (E) Excerpt of Sanger sequencing chromatogram showing the GAA motif of the expanded allele of patient 1.1.

Table summarizes the main features of the 9 patients with GAA-FGF14 ataxia. The median age at onset was 59 years (range, 28–67 years). We identified 1 patient who had an age at disease onset of 28 years, confirming that GAA-FGF14 ataxia can begin before the age of 30 years. All patients displayed a slowly progressive cerebellar syndrome, although none had dysarthria. A single patient experienced episodic symptoms, which were triggered by alcohol intake and stress, while 56% of patients (5/9) reported diplopia and 44% (4/9) reported vertigo and/or dizziness. Horizontal gaze-evoked nystagmus was observed in 78% of cases (7/9), consistent with previous reports showing that cerebellar oculomotor signs are common in GAA-FGF14 ataxia.^1,9 However, none of the 9 patients exhibited downbeat nystagmus. The patient carrying biallelic GAA repeat expansions had a relatively rapid disease progression (spinocerebellar degeneration functional score [SDFS] disability score¹⁰ of 3 after 3-year disease duration, indicating moderate gait impairment). In addition, extrapyramidal features were observed in 2 patients.

Overall, the median disease duration was 12 years (range, 3–39 years). Two patients required unilateral walking aid after 7 and 12 years of disease duration, respectively. One patient became wheelchair bound after long-standing disease (15 years), in keeping with previous reports showing that wheelchair dependence is uncommon in GAA-FGF14 ataxia.⁹ All other patients were still ambulating independently at the last follow-up, including 2 patients who had a disease duration exceeding 20 years.

In keeping with some reports showing weak to no significant correlation between size of FGF14 expansion and age at onset or disease severity,^1,7,9 we identified 3 patients who presented marked differences in age at onset and disease progression despite carrying an expansion of similar size (patient 2, 376 repeats; patient 3, 449 repeats; and patient 4, 424 repeats). Notably, patient 3 developed disease at age 28 years and had a more gradual disease progression (SDFS score of 1 after 23 years of disease duration) compared with patient 4 who developed disease at age 59 years and had an SDFS score of 3 after 9 years of disease duration.

One patient had evidence of mild axonal sensory polyneuropathy on electrophysiologic studies. Isolated vermis atrophy was observed in 4 patients (4/8; 50%) on brain MRI, while diffuse cerebellar atrophy was observed in 3 patients (3/8; 37%). Mild midbrain and pons atrophy was observed along with diffuse cortical atrophy in 1 patient. This patient had a comparatively more severe phenotype, with dystonia of the upper limbs, vestibular areflexia, and was wheelchair dependent. Although it is unknown whether these findings are pathologically related to GAA-FGF14 ataxia, an extensive workup did not reveal an alternative diagnosis.

Of the GAA-FGF14–negative patients, we identified 4 patients who carried a non-GAA expansion (2 (GAAGGA)_n, 1 (GAAGAAA)_n, and 1 [(GAA)₄(GCA)₁]_n). Non-GAA expansions in FGF14 are likely not pathogenic, as previously shown in segregation studies.¹