Assessment of the reliability, responsiveness, and meaningfulness of the scale for the assessment and rating of ataxia (SARA) for lysosomal storage disorders
Julien Park Tatiana Bremova-Ertl Mario Manto
Objective To evaluate the reliability, responsiveness, and validity of the Scale for the Assessment and Rating of Ataxia (SARA) in patients with lysosomal storage disorders (LSDs) who present with neurological symptoms, and quantify the threshold for a clinically meaningful change.
Methods We analyzed data from three clinical trial cohorts (IB1001-201, IB1001-202, and IB1001-301) of patients with Niemann–Pick disease type C (NPC) and GM2 Gangliosidoses (Tay–Sachs and Sandhoff disease) comprising 122 patients and 703 visits. Reproducibility was described as re-test reliability between repeat baseline visits or baseline and post-treatment washout visits. Responsiveness was determined in relation to the Investigator’s, Caregiver’s, and Patient’s Clinical Global Impression of Improvement (CGI-I). The CGI-I data was also used to quantify a threshold for a clinically meaningful improvement on the SARA scale. Using a qualitative methods approach, patient/caregiver interviews from the IB1001-301 trial were further used to assess a threshold of meaningful change as well as the breadth of neurological signs and symptoms captured and evaluated by the SARA scale.
Results The Inter-Class Correlation (ICC) was 0.95 or greater for all three trials, indicating a high internal consistency/reliability. The mean change in SARA between repeat baseline and post-treatment washout visit assessments in all trials was -0.05, SD 1.98, i.e., minimal, indicating no significant differences, learning effects or other systematic biases. For the CGI-I responses and change in SARA scores, Area Under the Curve (AUC) values were 0.82, 0.71, and 0.77 for the Investigator’s, Caregiver’s, and Patient’s CGI-I respectively, indicating strong agreement. Further qualitative analyses of the patient/caregiver interviews demonstrated a 1-point or greater change on SARA to be a clinically meaningful improvement which is directly relevant to the patient’s everyday functioning and quality of life. Changes captured by the SARA were also paralleled by improvement in a broad range of neurological signs and symptoms and beyond cerebellar ataxia.
Conclusion Qualitative and quantitative data demonstrate the reliability and responsiveness of the SARA score as a valid measure of neurological signs and symptoms in LSDs with CNS involvement, such as NPC and GM2 Gangliosidoses. A 1-point change represents a clinically meaningful transition reflecting the gain or loss of complex function.