Typical Friedreich's ataxia without GAA expansions and GAA expansions without typical Friedreich's ataxia

We clinically assessed and performed polymerase chain reaction analysis for the GAA trinucleotide repeat expansion in 103 patients from 73 families in Ireland, with a prior clinical diagnosis of Friedreich's ataxia (FA) or an unclassified progressive ataxic syndrome. The patients were classified as "typical" or "atypical" FA according to Harding's mandatory clinical diagno stic criteria. All patients underwent blood glucose analysis, and electroca rdiography and echocardiography was performed in 99 and 101 patients, respe ctively. Mutation screening for expanded CAG trinucleotide repeats, associa ted with spinocerebellar ataxia(SCA) 1, 2, 3 and 6 was performed in 86 pati ents overall, including all GAA negative patients. Forty-nine of 56 typical patients and 13 of 47 atypical patients were either homozygous or heterozy gous for the GAA expansion. Seven patients with a typical FA phenotype were negative for the GAA expansion. Although one of these patients had vitamin E deficiency, and two had raised a-fetoprotein levels, three other GAA neg ative patients with a typical FA phenotype had no other identifiable cause for their ataxia, once again raising the possibility of locus heterogeneity in FA. It is also possible that these patients have two point mutations in the X25 gene, or that they have another ataxic syndrome mimicking the FA p henotype. Two families who were homozygous for the GAA expansion exhibited intrafamilial phenotypic variability. Only one GAA negative patient had the SCA 3 mutation, and this was the only patient in the study with a possible autosomal dominant inheritance pattern. III the homozygous GAA population typical patients had significantly more repeats on the smaller allele than atypical patients, and there was an inverse relationship between the number of repeats on the smaller allele and the age at presentation. There was al so an inverse relationship between the repeat size on both the larger and t he smaller of the two alleles and the age at becoming wheelchair bound. The re was no significant relationship between repeat size and the other indice s of disease severity, including the presence or absence of diabetes or car diomyopathy. This is the first large study of an Irish population with prog ressive ataxia that has shown a similar phenotype/genotype relationship to studies of FA in other European and non-European populations, The relativel y low sensitivity and specificity of Harding's clinical diagnostic criteria must be appreciated when clinically assessing patients with a progressive ataxic syndrome. Although molecular genetic analysis now plays an essential role in diagnosis and classification, patients with a typical FA phenotype without any identifiable cause for their ataxia exist.