Delineating the pathogenic threshold and phenotypic spectrum of SCA27B: findings from a large French-Canadian cohort

Abstract

Background: Autosomal dominant spinocerebellar ataxia 27B (SCA27B), caused by an intronic (GAA•TTC) repeat expansion in FGF14, is a common cause of late-onset cerebellar ataxia, but its genotypic and phenotypic spectrum remains to be fully established.

Methods: We analysed the FGF14 (GAA•TTC) repeat expansion in a cohort of 134 patients with ataxia and 822 controls from Quebec. We conducted segregation study in large families to further characterize intergenerational repeat instability.

Results: We found a significant enrichment of (GAA•TTC)_≥200 alleles in the ataxia cohort compared to controls (53.0%, 71/134, vs 3.6%, 30/822, p < 0.0001), including for (GAA•TTC)_200-249 alleles (8.2% vs 2.6%, p = 0.0026). We identified 12 ataxic patients with a phenotype compatible with SCA27B carrying a (GAA•TTC)_200-249 expansion supporting the pathogenicity of these alleles in some patients. We further delineated the phenotype of 125 symptomatic individuals from 69 families who carried an FGF14 (GAA•TTC)_≥200 repeat expansion. Patients with (GAA•TTC)_200-249, (GAA•TTC)_250-299, and (GAA•TTC)_≥300 had a similar phenotype. We observed that 14% of patients with episodic symptoms (13/92) had severe episodes that were initially misdiagnosed as stroke, vestibular neuritis, Wernicke's encephalopathy, or seizures.

Discussion and conclusion: This large cohort demonstrates that (GAA•TTC)_200-249 alleles are enriched in patients with ataxia compared to controls and can be pathogenic for SCA27B, supporting the need to define a lower pathogenic threshold in the presence of specific clinical criteria.

Keywords: Ataxia; FGF14; French–Canadian; SCA27B.

Fig. 1

Allele distribution in the control and the ataxia cohorts. The frequency of (GAA•TTC)_200–249 (OR 3.41, 95% CI 1.56–7.29; Fisher´s exact test, p = 0.0026), (GAA•TTC)_250–299 (OR 12.51, 95% CI 4.73–32.25; Fisher´s exact test, p < 0.0001), and (GAA•TTC)_≥300 (OR 221.5, 95% CI 58.83–933.6; Fisher´s exact test, p < 0.0001) expansions was significantly higher in patients with ataxia than in controls. Non-GAA-pure expansions had a similar proportion in patients and controls (OR 0.68, 95% CI 0.06–4.14; Fisher´s exact test, p = 0.71)

Fig. 2

Intergenerational instability of the FGF14 (GAA•TTC) repeat locus. A) Analysis of GAA repeat size changes across 131 intergenerational transmissions, including 84 maternal and 47 paternal events. B) Change in GAA repeat length across 131 intergenerational transmissions. The y-axis shows the change in repeat length from parent to child. In A and B, alleles inherited from the mother are shown in blue and those from the father in orange. Contractions are plotted below the dashed lines while expansions are plotted above them

Fig. 3

Sizes of non-GAA-pure interruptions (blue) and (GAA•TTC) tracts (orange) in 28 alleles with interrupted FGF14 alleles. The alleles were derived from 25 individuals across five families. While the (GAA•TTC) repeat varied within families, the size of the non-GAA-pure interruptions remained stable

Fig. 4

Patients with (GAA•TTC)_200–249 expansions. 53 patients carrying a (GAA•TTC)_200–249 allele were identified and grouped according to their clinical status and phenotype (unaffected carriers, patients with a confirmed alternative diagnosis, patients with unsolved ataxia and a phenotype compatible with SCA27B, and patients with unsolved ataxia and a phenotype atypical for SCA27B). No statistically significant difference in the median repeat size was found between the four groups (Kruskal–Wallis test, p = 0.11). Affected and unaffected individuals from confirmed SCA27B families are shown in blue

Fig. 5

Family trees showing co-occurrence of affected individuals carrying (GAA•TTC)_200–249 alleles repeats (A.III.7, A.III.12, B.III.3, B.III.6), and alleles (GAA•TTC)_≥250. Some asymptomatic individuals also carry (GAA•TTC)_200–249 alleles (A.III.3, A.III.9, A.III.11) supporting the incomplete penetrance of such alleles. Age at last exam was noted for asymptomatic individuals carrying a 200–249 repeat allele

Fig. 6

Clinical characterization and disease progression in SCA27B. a Inverse correlation between age of onset and repeat size. Pearson’s correlation coefficient = −0.40, R² = 0.16, p < 0.0001. b Distribution of scores per SARA items represented as a fraction of the total SARA score. c Longitudinal intra-individual progression of ataxia severity as assessed by SARA relative to disease duration (50 observations from 21 patients). Observations from the same patient are connected by a dotted line. SARA score increased on average by 0.22 points/year of disease duration. The solid black line shows the average progression of the SARA score over disease duration across all patients as modelled by a linear mixed-effects model accounting for disease duration as fixed effect (p = 0.016). d Longitudinal intra-individual progression of functional impairment as assessed by SDFS score relative to disease duration (182 observations from 51 patients). Observations from the same patient are connected by a dotted line. SDFS score increased on average by 0.16 points/year of disease duration (linear mixed-effects model, p < 0.0001)

Fig. 7

Effect of sex on age at onset and disease progression in SCA27B a Median age at onset by sex showed no significant differences between females and males. Mann–Whitney U test: p = 0.98. b Relationship between repeat size and age at onset stratified by sex of the patient. c Longitudinal intra-individual progression of ataxia severity as assessed by SARA relative to disease duration in men and women. d Longitudinal intra-individual progression of functional impairment as assessed by SDFS score relative to disease duration in men and women