Best Oculomotor Endpoints for Clinical Trials in Hereditary Ataxias: A Systematic Review and Consensus by the Ataxia Global Initiative Working Group on Digital‑Motor Biomarkers.

Abstract

Oculomotor deficits are common in hereditary cerebellar ataxias (HCAs) and their quantitative assessment offers a sensitive and reliable manner to capture disease-severity and progression. As a group of experts of the Ataxia Global Initiative to support trial readiness, we previously established harmonized methodology for quantitative oculomotor assessments in HCAs. Here, we aimed to identify to most promising oculomotor/vestibular outcomes as endpoints for future trials. Through a systematic MEDLINE search we identified 130 articles reporting oculomotor/vestibular recordings in patients with HCAs. A total of 2,018 subjects were included: 1,776 with genetically-confirmed and 242 with clinically-defined HCAs. Studied diseases included spinocerebellar ataxias (SCA) 1/2/3/6/7/27B, episodic ataxia type 2, Friedreich ataxia, RFC1-related ataxia, fragile X-associated tremor/ataxia syndrome, cerebrotendinous xanthomatosis, ataxia-telangiectasia, ataxia with oculomotor apraxia types 1&2, and Niemann-Pick disease type C. We identified up to four oculomotor/vestibular outcomes per diagnostic entity, based on their ability to robustly discriminate patients from controls, correlate with disease-severity, detect longitudinal change, and represent different disease stages. For each parameter we provide recommendations for recordings. While the implementation of quantitative assessments into clinical trials offers a unique opportunity to track dysfunction of oculomotor/vestibular networks and to assess the impact of interventions, in some HCAs, endpoint qualification of available outcomes requires further validation to characterize their reliability, sensitivity to change, and minimally important change to patients. For all HCAs for which quantitative data are scarce or lacking, there is an urgent need for prospective studies covering a broader range of oculomotor/vestibular domains as approaching new treatments require harmonized and reliable endpoints.