Age at onset variance analysis in spinocerebellar ataxias: a study in a Dutch-French cohort.

Abstract

In dominant spinocerebellar ataxias (SCAs), the issue of whether non-CAG dependent factors contribute to onset age remains unsettled. Data on SCA genotype, onset age, normal/expanded CAG repeat length, sex of the patient and transmitting parent, and family details were available from 802 patients. Based on the model [log(10) (age at onset) = k - b CAG(exp) + epsilon], we examined changes in adjusted R(2) and residual standard error following incorporation of the other factors in this model. The expanded repeat explained 44.3 to 74.9% of onset age variance, although this was less than 50% in SCA3 and SCA6, implicating a large effect of non-CAG factors. The relation between onset age and CAG repeat was similar for SCA1, 3, 6, and 7, but different for SCA2, pointing to different polyglutamine effects in SCA2. For SCA2 and SCA3, 17.1 and 45.5% of onset age variance, respectively, were explained by currently (unidentified) familial factors. We found a significant contribution of the nonexpanded allele in SCA1 and SCA6. Besides polyglutamine motif (determined by the expanded CAG repeat length), we identified the following age at onset modifiers: protein context in SCA2; familial factors in SCA2 and SCA3; and the nonexpanded CAG repeat in SCA1 and SCA6.