Longitudinal analysis of the neurological features of ataxia-telangiectasia

AIM To assess the relationship between genotype and neurological progression in ataxia-telangiectasia (A-T). METHODS Clinical and laboratory data were extracted retrospectively from the records of patients attending the UK National Ataxia-Telangiectasia Clinic. Neurological assessments were performed using the A-T Index (Crawford Score) and the A-T Neurological Examination Scale Toolkit (A-T NEST). Variables influencing phenotype were identified by using an information-theoretic approach starting from a maximal model to generate estimates of coefficients for each variable. Per-individual progression was assessed for patients with three or more clinic attendances. RESULTS The genotype could be determined for 125/135 patients. Crawford and A-T NEST scores were well correlated. For both scoring systems the estimated coefficients were significantly positive for Age x kinase activity but not Age x protein expression. Unlike the per-genotype analysis, the individual progression of neurological scores in the 34 patients that attended on three or more occasions was not smooth and linear (and in some cases improved over time). INTERPRETATION Residual kinase activity confers a milder phenotype but there is no difference between kinase-dead and protein-null genotypes. The non-linear progression of individual patients' neurological scores may reflect biological complexity, day-to-day variability, limitations of the assessment methods or a combination of all three.